Phase 2 - EDA and Sentiment Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "66f11742",
   "metadata": {},
   "source": [
    "# DATA SCIENCE CAREER GUIDANCE SYSTEM"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84b02798",
   "metadata": {},
   "source": [
    "In this project the data is scrapped from simpyhired.com for Data Science jobs and careerbliss.com for respective Company reviews which is intregrrated together in a single dataset stored in csv format. The data collected is used to analyze the job description and company reviews through sentiment analysis to give a recommendation to the users about probalility of their career growth if they join the particular company"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c5e5c8da",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Raghul Devaraj\\anaconda3\\envs\\env_dlib\\lib\\site-packages\\scipy\\__init__.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.23.3\n",
      "  warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "\n",
    "#library for text pre-processing\n",
    "import string\n",
    "string.punctuation\n",
    "import re\n",
    "import collections\n",
    "import nltk\n",
    "from nltk.tokenize import word_tokenize\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.stem.porter import PorterStemmer\n",
    "from nltk.stem import WordNetLemmatizer\n",
    "from collections import Counter\n",
    "from nltk.util import ngrams\n",
    "import datetime as dt\n",
    "from datetime import date\n",
    "import calendar\n",
    "\n",
    "# Libraries for Sentiment Analysis\n",
    "from nltk.sentiment.vader import SentimentIntensityAnalyzer\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "53ec6370",
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option(\"display.max_colwidth\", -1)\n",
    "df = pd.read_csv('engine_x_final.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "de025b54",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(['Unnamed: 0', 'Unnamed: 0.1'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1fdebcb4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 390 entries, 0 to 389\n",
      "Data columns (total 18 columns):\n",
      " #   Column                Non-Null Count  Dtype  \n",
      "---  ------                --------------  -----  \n",
      " 0   Job_title             390 non-null    object \n",
      " 1   company_name          390 non-null    object \n",
      " 2   company_location      390 non-null    object \n",
      " 3   job_type              325 non-null    object \n",
      " 4   posted_date           143 non-null    object \n",
      " 5   qualification         390 non-null    object \n",
      " 6   benefits              390 non-null    object \n",
      " 7   description           390 non-null    object \n",
      " 8   average_rating_list   172 non-null    float64\n",
      " 9   company_culture       172 non-null    float64\n",
      " 10  growth_opportunities  172 non-null    float64\n",
      " 11  people_you_work_with  172 non-null    float64\n",
      " 12  person_you_work_for   172 non-null    float64\n",
      " 13  rewards_you_receive   172 non-null    float64\n",
      " 14  suppport_you_get      172 non-null    float64\n",
      " 15  way_you_work          172 non-null    float64\n",
      " 16  work_setting          172 non-null    float64\n",
      " 17  reviews               176 non-null    object \n",
      "dtypes: float64(9), object(9)\n",
      "memory usage: 55.0+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bdb3aa36",
   "metadata": {},
   "outputs": [],
   "source": [
    "categorical_features = [feature for feature in df.columns if df[feature].dtypes == 'O']\n",
    "numerical_features = [feature for feature in df.columns if df[feature].dtypes != 'O']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca04af73",
   "metadata": {},
   "source": [
    "Number of observations = 390\n",
    "\n",
    "Number of features = 18\n",
    "\n",
    "Number of Numerical features = 9\n",
    "\n",
    "Number of categorical features = 9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c14cbdba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Job_title</th>\n",
       "      <th>company_name</th>\n",
       "      <th>company_location</th>\n",
       "      <th>job_type</th>\n",
       "      <th>posted_date</th>\n",
       "      <th>qualification</th>\n",
       "      <th>benefits</th>\n",
       "      <th>description</th>\n",
       "      <th>average_rating_list</th>\n",
       "      <th>company_culture</th>\n",
       "      <th>growth_opportunities</th>\n",
       "      <th>people_you_work_with</th>\n",
       "      <th>person_you_work_for</th>\n",
       "      <th>rewards_you_receive</th>\n",
       "      <th>suppport_you_get</th>\n",
       "      <th>way_you_work</th>\n",
       "      <th>work_setting</th>\n",
       "      <th>reviews</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DATA SCIENTIST</td>\n",
       "      <td>PUMA</td>\n",
       "      <td>Bengaluru, Karnataka</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2 days ago</td>\n",
       "      <td>['Azure', 'SQL', 'Spark', 'Performance marketing', 'Deep learning', \"Bachelor's degree\", \"Master's degree\"]</td>\n",
       "      <td>Not mentioned</td>\n",
       "      <td>INFORMATION TECHNOLOGY\\n\\nDATA SCIENTIST\\n\\nBengaluru, India\\n\\nSPEED &amp; SPIRIT is what we look for in our candidates, defined by some simple values that inspire us to BE DRIVEN in our performance, BE VIBRANT in our sporting legacy, BE TOGETHER in our team spirit, and BE YOU to let our individual talent and experience shine. Applying for a job at PUMA is easy and all genders are welcome. Simply click APPLY ONLINE and follow the steps to upload your application.\\n\\nYOUR MISSION\\n\\nPosition Summary &amp; Key Objectives\\n\\nPuma Data Science and Analytics team is looking for Data scientists to join in Bangalore HQ. These Data Scientists will help us analyze TBs of data, identify trends, help us make smarter decisions to keep us FOREVER FASTER. Primary focus for DSs will be in applying data mining techniques, doing statistical analysis, building business insights and high-quality automated algorithms, and end-to-end data products. The role requires high technical aptitude and problem-solving abilities, motivation, intellectual curiosity, and a passion for data discovery, analytics, data visualization and, in general, how we can improve different aspects of the business. Data Scientists will help us in areas of demand planning and forecasting, pricing, performance marketing, CRM, user Personalization, Search and Discovery etc.\\n\\nThe Data Science team is a business-critical team that is at the center of innovation for the company and is a major driver of company's products and business\\n\\nThe opportunity to work with a world-class team of talented business analysts, data engineers, business heads with the strong background, and who are extremely passionate about solving real-world applied problems in the retail and fashion industry\\n\\nA real opportunity to expand your skill-sets and a chance to learn and use the latest advances and approaches in data science and data mining on real-world data-sets\\n\\nWorking at a fast-paced work environment where you are expected to be self-motivated and are measured on your performance\\n\\nWork as a data scientist, strategist and evangelist, identifying and integrating new datasets that can be leveraged and work closely with the business teams to strategize and execute the development of data products Collaborate with business, engineering and product teams to understand company needs and devise possible solutions to streamline the delivery Implement new statistical or other mathematical methodologies as needed for specific models or analysis Creating end to end automated and integrated systems and constant tracking of its performance and accuracy Execute analytical experiments methodically to help solve various problems and make a true impact across various functions Implement data science models into production by collaborating with software developers and data engineers. Communicate analytic solutions to business stakeholders, Function heads and implement improvements as needed to operational systems\\n\\nYOUR TALENT\\n\\nQualifications and Functional Competencies\\n\\nJob Responsibility:\\nAbility to understand a problem statement and implement analytical solutions &amp; techniques Work and collaborate with other teams to deliver the result best suited to drive revenue of the company Fast learner: ability to learn and pick up a new language/tool/ platform quickly Conceptualize, design and deliver high-quality solutions and insightful analysis Conduct research and prototyping innovations; data and requirements gathering; solution scoping and building solution architecture\\n\\nExperience Required:\\n1. Expert level proficiency in SQL, R and Python, Spark with 1-2 years of experience\\n2. Ability to discover efficient solutions to complex problems\\n\\nExperience with data processing, database programming and data analytics, data sciences Deep understanding of statistical modeling (Regression, Clustering, Decision trees), machine learning, algorithms, data mining concepts, and a track record of solving problems with these methods Experience with Gradient Descent, Clustering, Deep Learning, SVM, k-nearest neighbors, random forests, ensemble methods, Logistic Regression, Probability Theory, Bayesian/Markov networks. Demonstrated programming skills in relevant languages, including R, Python, PySpark Optional Skills: Azure, AWS, Databricks, MLFlow Problem-solving: Ability to break the problem into small problems and think of relevant techniques which can be explored &amp; used to cater to those\\n\\nIntermediate to advanced knowledge of machine learning, probability theory, statistics and algorithms. You will be required to discuss and use various algorithms and approaches on daily basis\\n\\nEducation Qualification:\\nB.Tech / M.Tech in Computer Science / Mathematics / Signal Processing or related fields from one of the premier Institutes (IITs/NITs/BITS)\\n\\nPUMA provides equal opportunities for all job applicants, regardless of race, color, religion, national origin, sex, gender identity or expression, sexual orientation, age, or disability. Equality for all is one of the core principles at PUMA and we do not tolerate any form of harassment or discrimination.\\n\\nPUMA supports over 16,100 employees across 120+ countries. The PUMA Group owns the brand PUMA, Cobra Golf and stichd, and is headquartered in Herzogenaurach, Germany.</td>\n",
       "      <td>3.6</td>\n",
       "      <td>3.3</td>\n",
       "      <td>3.3</td>\n",
       "      <td>4.5</td>\n",
       "      <td>3.5</td>\n",
       "      <td>3.5</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Ensure promotions are given to people who deserve it, change the loyalty gathering process, and make sure all higher ups work in teh field so they can know what works in reality and what does not.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Associate Architect - Fullstack (Python/ Ruby and Angular)</td>\n",
       "      <td>Clarivate</td>\n",
       "      <td>Bengaluru, Karnataka</td>\n",
       "      <td>Full-time</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['React', 'Continuous integration', 'SQL', '.NET', 'CI/CD', \"Bachelor's degree\", \"Master's degree\"]</td>\n",
       "      <td>Not mentioned</td>\n",
       "      <td>We are looking for an Associate Software Architect with a history of design and build a comprehensive data analytic system.\\n\\nWorking with both business owners and development teams, your responsibilities include architect, lead, design, develop and deliver an end-to-end data analytic platform consist of data entry, data enrichment, data mastering, and data analytic systems.\\n\\nResponsibilities:\\nParticipate in all phases of the software development lifecycle, design and implement elegant solutions to enhance our existing data analytic systems.\\nLead the understanding of the business requirements to ensure optimal solution design aligned to enterprise architecture principles and best practices, then support the implementation of those designs to the development teams.\\nCollaborate with business teams, development teams, and QA teams to organize the product roadmap, plan short-term and long-term goals, and forecast release timelines within an Agile environment.\\nCreate, revise, and maintain design/technical documentation.\\nCoach and help the development teams on best practices and stay up-to-date with industry-leading practices and technologies.\\nLiaise with business teams and development teams in evaluating project constraints, proposing alternatives, alleviate risks and perform reengineering when warranted.\\nProduce efficient, maintainable, and well-tested code, by practicing code review process, following code style guides, writing documentation, and ensuring test coverage.\\nParticipate in the SCRUM ceremonies.\\n\\nQualifications:\\nGraduates in BE/BTech/MTech – Computer science\\n10+ years of software engineering experience\\n5+ years of experience in solution design and development\\nExpert knowledge with Service Oriented Architecture design and development\\nStrong knowledge of Frontend Frameworks such as AngularJs, Ember, and React\\nStrong knowledge of Backend Development skills in Python, Java, Ruby or C#\\nSolid Understanding of Database and SQL\\nExtensive experience with software development best practices, including version control, documentation, testing, and CI/CD\\nPreferably having knowledge in DW and ETL\\nStrong organizational, communication, analytical, troubleshooting, and problem-solving skills\\n\\nIt is the policy of Clarivate to provide equal employment opportunity (EEO) to all persons regardless of age, color, national origin, citizenship status, physical or mental disability, race, religion, creed, gender, pregnancy, sex, sexual orientation, gender identity and/or expression, genetic information, marital status, status with regard to public assistance, veteran status, or any other characteristic protected by federal, state or local law. In addition, Clarivate will provide reasonable accommodations for qualified individuals with disabilities.</td>\n",
       "      <td>4.8</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Really great company who puts their employees first. Everyone is so friendly and helpful to their coworkers. The company is great at what it does, helping libraries to access the content they'd like to make available to their patrons. My colleagues were very talented and helped me to build a skillset in the IT field.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Software Engineer</td>\n",
       "      <td>Dell Technologies</td>\n",
       "      <td>India</td>\n",
       "      <td>Full-time</td>\n",
       "      <td>8 days ago</td>\n",
       "      <td>['DNS', 'VMWare', 'ESXi', 'Software troubleshooting', 'TCP']</td>\n",
       "      <td>Not mentioned</td>\n",
       "      <td>Dell’s IEO (Infrastructure Engineering &amp; Operations) organization provides a set of shared services for engineering labs and engineering teams across all divisions of ISG, supporting product development. The Platform Infrastructure &amp; Orchestration Engineering team within IEO has team members based around the globe and is responsible for delivering infrastructure and platform services as well as automation for multiple cloud platforms across engineering labs. Our goal is to provide enterprise-level services to all ISG Engineering users/consumers as well as many lab managers and application owners in the DevOps ecosystem.\\n\\nJoin us as a Software Engineer on our Engineering Development team in India to do the best work of your career and make a profound social impact.\\n\\nWhat you’ll achieve\\nAs a Software Engineer, you will be responsible for developing sophisticated systems and software basis the customer’s business goals, needs and general business environment creating software solutions.\\n\\nYou will:\\nAs part of a team, you will help to review and set requirements, as well as build, tune, and maintain on-prem VMware Cloud infrastructure in an enterprise global engineering environmentYou will take an active role in maintaining system build documentation, tuning system performance, troubleshooting, and installing system-wide software to maintain current and future standards for the environmentYou will assist support teams in troubleshooting issues as requiredYou will help to monitor the health of the platforms and make recommendations for additional allocation of computing and storage resources to meet forecasted demand\\nTake the first step towards your dream career\\nEvery Dell Technologies team member brings something unique to the table. Here’s what we are looking for with this role:\\n\\nEssential Requirements\\n\\nMust have an understanding of virtualization technologies e.g., VMWare ESX, NSX, vSphere 6.5/7.0, vRealize Operations. This includes installation, configuration, and updating ESXi hosts on a regular basisProficiency with VMWare networking and storage solutions: Distributed virtual switches, vSAN, NSX-T, DNS, TCP/IPExperience with scripting and automation including PowerShell/PowerCLI or similar toolsPrior experience as a Systems Administrator, Solutions Engineer, Software Engineer, Systems Engineer, or Technology Consultant a plusRequires a degree in Science, Technology, Engineering or Mathematics (STEM), or related STEM degree program from an Accredited University/School/College or equivalent on the job experience, with at least 1-3 years of related work experience\\nHere’s our story; now tell us yours\\n\\nDell Technologies helps organizations and individuals build a brighter digital tomorrow. Our company is made up of more than 150,000 people, located in over 180 locations around the world. We’re proud to be a diverse and inclusive team and have an endless passion for our mission to drive human progress.\\n\\nWhat’s most important to us is that you are respected, feel like you can be yourself, and have the opportunity to do the best work of your life - while still having a life. We offer excellent benefits, bonus programs, flexible work arrangements, a variety of career development opportunities, employee resource groups, and much more.\\n\\nWe started with computers, but we didn’t stop there. We are helping customers move into the future with multi-cloud, AI, and machine learning through the most innovative technology and services portfolio for the data era. Join us and become a part of what’s next in technology, starting today.\\n\\nYou can also learn more about us by reading our latest Diversity and Inclusion Report and our plan to make the world a better place by 2030 here.\\n\\nApplication closing date: 30 October 2022\\n\\nDell is committed to the principle of equal employment opportunity for all employees and to providing employees with a work environment free of discrimination and harassment. All employment decisions at Dell are based on business needs, job requirements and individual qualifications, without regard to race, color, religion or belief, national, social or ethnic origin, sex (including pregnancy), age, physical, mental or sensory disability, HIV Status, sexual orientation, gender identity and/or expression, marital, civil union or domestic partnership status, past or present military service, family medical history or genetic information, family or parental status, or any other status protected by the laws or regulations in the locations where we operate. Dell will not tolerate discrimination or harassment based on any of these characteristics. Dell encourages applicants of all ages. Read the full Equal Employment Opportunity Policy here.</td>\n",
       "      <td>3.9</td>\n",
       "      <td>3.8</td>\n",
       "      <td>3.4</td>\n",
       "      <td>4.3</td>\n",
       "      <td>4.1</td>\n",
       "      <td>3.9</td>\n",
       "      <td>3.8</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.9</td>\n",
       "      <td>Don't be so afraid of coming up with new things and seeing what the development staff can do! If we hadn't broken a few of their IT mandates in 1995, there would have been no support.dell.com worth talking about!!!! Also, don't have all of the developers in IT and crushed by budgets and IT timelines - get them out into the business units so they can be more effective and react faster to changing markets!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>#RestartWithInfosys-EAIS-Java/Springboot/Microservices Consultant</td>\n",
       "      <td>Infosys Limited</td>\n",
       "      <td>Bengaluru, Karnataka</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21 hours ago</td>\n",
       "      <td>['APIs', 'Microservices', 'Java', \"Bachelor's degree\", \"Master's degree\"]</td>\n",
       "      <td>Not mentioned</td>\n",
       "      <td>Responsibilities\\nThis Program is for Women only! Women professionals who are on a career break, currently. These women should have worked for at least 2 years before taking a career break and should have a minimum of 6 months (continuous) break in their career. A day in the life of an Infoscion • As part of the Infosys consulting team, your primary role would be to actively aid the consulting team in different phases of the project including problem definition, effort estimation, diagnosis, solution generation and design and deployment • You will explore the alternatives to the recommended solutions based on research that includes literature surveys, information available in public domains, vendor evaluation information, etc. and build POCs • You will create requirement specifications from the business needs, define the to-be-processes and detailed functional designs based on requirements. • You will support configuring solution requirements on the products; understand if any issues, diagnose the root-cause of such issues, seek clarifications, and then identify and shortlist solution alternatives • You will also contribute to unit-level and organizational initiatives with an objective of providing high quality value adding solutions to customers. If you think you fit right in to help our clients navigate their next in their digital transformation journey, this is the place for you!\\n\\nAdditional Responsibilities:\\nAbility to work with clients to identify business challenges and contribute to client deliverables by refining, analyzing, and structuring relevant data • Awareness of latest technologies and trends • Logical thinking and problem solving skills along with an ability to collaborate • Ability to assess the current processes, identify improvement areas and suggest the technology solutions • One or two industry domain knowledge\\nTechnical and Professional Requirements:\\nPrimary skills:Cloud Platform-&gt;Microservices,Java,Java-&gt;Java API,Java-&gt;Microservices,Java-ALL-&gt;Java Cloud,Springboot\\nPreferred Skills:\\nCloud Platform-&gt;Microservices,Java-&gt;Microservices,Springboot,Java-ALL-&gt;Java Cloud,Java-&gt;Java API,Java\\nEducational Requirements\\nBachelor of Engineering,Bachelor Of Technology,Bachelor Of Science (Engg),Master Of Comp. Applications,Master Of Technology,MSc,BCA,BCom,MCOM</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.8</td>\n",
       "      <td>3.7</td>\n",
       "      <td>4.4</td>\n",
       "      <td>4.2</td>\n",
       "      <td>3.5</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.2</td>\n",
       "      <td>4.0</td>\n",
       "      <td>I have worked for Infosys for 10 years in various roles and at various locations. The company has a wealth of resources which can be its strength when you are starting in the company. The large size of the company is a drawback for individual growth once an employee reaches middle-management in the company. I like to work in an environment where processes aid the decision. Having a large company with multiple hierarchy levels though drowns the work into processes and slow down productivity.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Data Annotator</td>\n",
       "      <td>Conxai Technologies GmbH</td>\n",
       "      <td>Bengaluru, Karnataka</td>\n",
       "      <td>Full-time</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Deep learning', 'Software deployment', 'PyTorch', 'Python', 'Data science', \"Bachelor's degree\"]</td>\n",
       "      <td>Not mentioned</td>\n",
       "      <td>We are building the first NO-CODE AI PLATFORM for the AEC industry, with the 3rd generation Explainable AI, that allows users to create complex use cases with zero coding at the frontend and neuro-symbolic AI under the hood.\\n\\nOur NO-CODE AI PLATFORM provides end-to-end functionalities starting with specifying knowledge, to creating and annotating datasets, all the way to training and deploying ML models.\\n\\nTasks\\n\\nWe are looking for a Data Scientist who will work with our AI Modelling Team in improving the accuracy and quality of our AI models to achieve excellent performance across several use cases.\\n\\nYou will be responsible for understanding customer use cases, translating these use cases into data requirements, annotating data, training, testing and deployment of AI models.\\n\\nA substantial part of your work will involve annotating data.\\n\\nFurthermore, you will be trained to work with our No-Code AI platform to train, evaluate and deploy AI Models across several use cases - all this without writing a single line of code.\\n\\nYou will also be contributing to the improvement of these AI Models using the python/pytorch programming language.\\n\\nThe candidate will gain unique professional training and experience as a No Code AI Data Specialist.\\n\\nRequirements\\n\\nBCA / BSc. / BTech or higher with experience in data annotation and data scienceExperience in working with deep learning models using pytorch.Experience with annotation, training, testing and deployment of AI models.Ability to support several projects at one timeSelf-motivated to learn about AI pipelines and processesFluent in English writing and speakingExcellent in communicationTeam player\\n</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                           Job_title  \\\n",
       "0  DATA SCIENTIST                                                      \n",
       "1  Associate Architect - Fullstack (Python/ Ruby and Angular)          \n",
       "2  Software Engineer                                                   \n",
       "3  #RestartWithInfosys-EAIS-Java/Springboot/Microservices Consultant   \n",
       "4  Data Annotator                                                      \n",
       "\n",
       "               company_name      company_location   job_type   posted_date  \\\n",
       "0  PUMA                      Bengaluru, Karnataka  NaN        2 days ago     \n",
       "1  Clarivate                 Bengaluru, Karnataka  Full-time  NaN            \n",
       "2  Dell Technologies         India                 Full-time  8 days ago     \n",
       "3  Infosys Limited           Bengaluru, Karnataka  NaN        21 hours ago   \n",
       "4  Conxai Technologies GmbH  Bengaluru, Karnataka  Full-time  NaN            \n",
       "\n",
       "                                                                                                 qualification  \\\n",
       "0  ['Azure', 'SQL', 'Spark', 'Performance marketing', 'Deep learning', \"Bachelor's degree\", \"Master's degree\"]   \n",
       "1  ['React', 'Continuous integration', 'SQL', '.NET', 'CI/CD', \"Bachelor's degree\", \"Master's degree\"]           \n",
       "2  ['DNS', 'VMWare', 'ESXi', 'Software troubleshooting', 'TCP']                                                  \n",
       "3  ['APIs', 'Microservices', 'Java', \"Bachelor's degree\", \"Master's degree\"]                                     \n",
       "4  ['Deep learning', 'Software deployment', 'PyTorch', 'Python', 'Data science', \"Bachelor's degree\"]            \n",
       "\n",
       "        benefits  \\\n",
       "0  Not mentioned   \n",
       "1  Not mentioned   \n",
       "2  Not mentioned   \n",
       "3  Not mentioned   \n",
       "4  Not mentioned   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                  description  \\\n",
       "0  INFORMATION TECHNOLOGY\\n\\nDATA SCIENTIST\\n\\nBengaluru, India\\n\\nSPEED & SPIRIT is what we look for in our candidates, defined by some simple values that inspire us to BE DRIVEN in our performance, BE VIBRANT in our sporting legacy, BE TOGETHER in our team spirit, and BE YOU to let our individual talent and experience shine. Applying for a job at PUMA is easy and all genders are welcome. Simply click APPLY ONLINE and follow the steps to upload your application.\\n\\nYOUR MISSION\\n\\nPosition Summary & Key Objectives\\n\\nPuma Data Science and Analytics team is looking for Data scientists to join in Bangalore HQ. These Data Scientists will help us analyze TBs of data, identify trends, help us make smarter decisions to keep us FOREVER FASTER. Primary focus for DSs will be in applying data mining techniques, doing statistical analysis, building business insights and high-quality automated algorithms, and end-to-end data products. The role requires high technical aptitude and problem-solving abilities, motivation, intellectual curiosity, and a passion for data discovery, analytics, data visualization and, in general, how we can improve different aspects of the business. Data Scientists will help us in areas of demand planning and forecasting, pricing, performance marketing, CRM, user Personalization, Search and Discovery etc.\\n\\nThe Data Science team is a business-critical team that is at the center of innovation for the company and is a major driver of company's products and business\\n\\nThe opportunity to work with a world-class team of talented business analysts, data engineers, business heads with the strong background, and who are extremely passionate about solving real-world applied problems in the retail and fashion industry\\n\\nA real opportunity to expand your skill-sets and a chance to learn and use the latest advances and approaches in data science and data mining on real-world data-sets\\n\\nWorking at a fast-paced work environment where you are expected to be self-motivated and are measured on your performance\\n\\nWork as a data scientist, strategist and evangelist, identifying and integrating new datasets that can be leveraged and work closely with the business teams to strategize and execute the development of data products Collaborate with business, engineering and product teams to understand company needs and devise possible solutions to streamline the delivery Implement new statistical or other mathematical methodologies as needed for specific models or analysis Creating end to end automated and integrated systems and constant tracking of its performance and accuracy Execute analytical experiments methodically to help solve various problems and make a true impact across various functions Implement data science models into production by collaborating with software developers and data engineers. Communicate analytic solutions to business stakeholders, Function heads and implement improvements as needed to operational systems\\n\\nYOUR TALENT\\n\\nQualifications and Functional Competencies\\n\\nJob Responsibility:\\nAbility to understand a problem statement and implement analytical solutions & techniques Work and collaborate with other teams to deliver the result best suited to drive revenue of the company Fast learner: ability to learn and pick up a new language/tool/ platform quickly Conceptualize, design and deliver high-quality solutions and insightful analysis Conduct research and prototyping innovations; data and requirements gathering; solution scoping and building solution architecture\\n\\nExperience Required:\\n1. Expert level proficiency in SQL, R and Python, Spark with 1-2 years of experience\\n2. Ability to discover efficient solutions to complex problems\\n\\nExperience with data processing, database programming and data analytics, data sciences Deep understanding of statistical modeling (Regression, Clustering, Decision trees), machine learning, algorithms, data mining concepts, and a track record of solving problems with these methods Experience with Gradient Descent, Clustering, Deep Learning, SVM, k-nearest neighbors, random forests, ensemble methods, Logistic Regression, Probability Theory, Bayesian/Markov networks. Demonstrated programming skills in relevant languages, including R, Python, PySpark Optional Skills: Azure, AWS, Databricks, MLFlow Problem-solving: Ability to break the problem into small problems and think of relevant techniques which can be explored & used to cater to those\\n\\nIntermediate to advanced knowledge of machine learning, probability theory, statistics and algorithms. You will be required to discuss and use various algorithms and approaches on daily basis\\n\\nEducation Qualification:\\nB.Tech / M.Tech in Computer Science / Mathematics / Signal Processing or related fields from one of the premier Institutes (IITs/NITs/BITS)\\n\\nPUMA provides equal opportunities for all job applicants, regardless of race, color, religion, national origin, sex, gender identity or expression, sexual orientation, age, or disability. Equality for all is one of the core principles at PUMA and we do not tolerate any form of harassment or discrimination.\\n\\nPUMA supports over 16,100 employees across 120+ countries. The PUMA Group owns the brand PUMA, Cobra Golf and stichd, and is headquartered in Herzogenaurach, Germany.   \n",
       "1  We are looking for an Associate Software Architect with a history of design and build a comprehensive data analytic system.\\n\\nWorking with both business owners and development teams, your responsibilities include architect, lead, design, develop and deliver an end-to-end data analytic platform consist of data entry, data enrichment, data mastering, and data analytic systems.\\n\\nResponsibilities:\\nParticipate in all phases of the software development lifecycle, design and implement elegant solutions to enhance our existing data analytic systems.\\nLead the understanding of the business requirements to ensure optimal solution design aligned to enterprise architecture principles and best practices, then support the implementation of those designs to the development teams.\\nCollaborate with business teams, development teams, and QA teams to organize the product roadmap, plan short-term and long-term goals, and forecast release timelines within an Agile environment.\\nCreate, revise, and maintain design/technical documentation.\\nCoach and help the development teams on best practices and stay up-to-date with industry-leading practices and technologies.\\nLiaise with business teams and development teams in evaluating project constraints, proposing alternatives, alleviate risks and perform reengineering when warranted.\\nProduce efficient, maintainable, and well-tested code, by practicing code review process, following code style guides, writing documentation, and ensuring test coverage.\\nParticipate in the SCRUM ceremonies.\\n\\nQualifications:\\nGraduates in BE/BTech/MTech – Computer science\\n10+ years of software engineering experience\\n5+ years of experience in solution design and development\\nExpert knowledge with Service Oriented Architecture design and development\\nStrong knowledge of Frontend Frameworks such as AngularJs, Ember, and React\\nStrong knowledge of Backend Development skills in Python, Java, Ruby or C#\\nSolid Understanding of Database and SQL\\nExtensive experience with software development best practices, including version control, documentation, testing, and CI/CD\\nPreferably having knowledge in DW and ETL\\nStrong organizational, communication, analytical, troubleshooting, and problem-solving skills\\n\\nIt is the policy of Clarivate to provide equal employment opportunity (EEO) to all persons regardless of age, color, national origin, citizenship status, physical or mental disability, race, religion, creed, gender, pregnancy, sex, sexual orientation, gender identity and/or expression, genetic information, marital status, status with regard to public assistance, veteran status, or any other characteristic protected by federal, state or local law. In addition, Clarivate will provide reasonable accommodations for qualified individuals with disabilities.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                        \n",
       "2  Dell’s IEO (Infrastructure Engineering & Operations) organization provides a set of shared services for engineering labs and engineering teams across all divisions of ISG, supporting product development. The Platform Infrastructure & Orchestration Engineering team within IEO has team members based around the globe and is responsible for delivering infrastructure and platform services as well as automation for multiple cloud platforms across engineering labs. Our goal is to provide enterprise-level services to all ISG Engineering users/consumers as well as many lab managers and application owners in the DevOps ecosystem.\\n\\nJoin us as a Software Engineer on our Engineering Development team in India to do the best work of your career and make a profound social impact.\\n\\nWhat you’ll achieve\\nAs a Software Engineer, you will be responsible for developing sophisticated systems and software basis the customer’s business goals, needs and general business environment creating software solutions.\\n\\nYou will:\\nAs part of a team, you will help to review and set requirements, as well as build, tune, and maintain on-prem VMware Cloud infrastructure in an enterprise global engineering environmentYou will take an active role in maintaining system build documentation, tuning system performance, troubleshooting, and installing system-wide software to maintain current and future standards for the environmentYou will assist support teams in troubleshooting issues as requiredYou will help to monitor the health of the platforms and make recommendations for additional allocation of computing and storage resources to meet forecasted demand\\nTake the first step towards your dream career\\nEvery Dell Technologies team member brings something unique to the table. Here’s what we are looking for with this role:\\n\\nEssential Requirements\\n\\nMust have an understanding of virtualization technologies e.g., VMWare ESX, NSX, vSphere 6.5/7.0, vRealize Operations. This includes installation, configuration, and updating ESXi hosts on a regular basisProficiency with VMWare networking and storage solutions: Distributed virtual switches, vSAN, NSX-T, DNS, TCP/IPExperience with scripting and automation including PowerShell/PowerCLI or similar toolsPrior experience as a Systems Administrator, Solutions Engineer, Software Engineer, Systems Engineer, or Technology Consultant a plusRequires a degree in Science, Technology, Engineering or Mathematics (STEM), or related STEM degree program from an Accredited University/School/College or equivalent on the job experience, with at least 1-3 years of related work experience\\nHere’s our story; now tell us yours\\n\\nDell Technologies helps organizations and individuals build a brighter digital tomorrow. Our company is made up of more than 150,000 people, located in over 180 locations around the world. We’re proud to be a diverse and inclusive team and have an endless passion for our mission to drive human progress.\\n\\nWhat’s most important to us is that you are respected, feel like you can be yourself, and have the opportunity to do the best work of your life - while still having a life. We offer excellent benefits, bonus programs, flexible work arrangements, a variety of career development opportunities, employee resource groups, and much more.\\n\\nWe started with computers, but we didn’t stop there. We are helping customers move into the future with multi-cloud, AI, and machine learning through the most innovative technology and services portfolio for the data era. Join us and become a part of what’s next in technology, starting today.\\n\\nYou can also learn more about us by reading our latest Diversity and Inclusion Report and our plan to make the world a better place by 2030 here.\\n\\nApplication closing date: 30 October 2022\\n\\nDell is committed to the principle of equal employment opportunity for all employees and to providing employees with a work environment free of discrimination and harassment. All employment decisions at Dell are based on business needs, job requirements and individual qualifications, without regard to race, color, religion or belief, national, social or ethnic origin, sex (including pregnancy), age, physical, mental or sensory disability, HIV Status, sexual orientation, gender identity and/or expression, marital, civil union or domestic partnership status, past or present military service, family medical history or genetic information, family or parental status, or any other status protected by the laws or regulations in the locations where we operate. Dell will not tolerate discrimination or harassment based on any of these characteristics. Dell encourages applicants of all ages. Read the full Equal Employment Opportunity Policy here.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                  \n",
       "3  Responsibilities\\nThis Program is for Women only! Women professionals who are on a career break, currently. These women should have worked for at least 2 years before taking a career break and should have a minimum of 6 months (continuous) break in their career. A day in the life of an Infoscion • As part of the Infosys consulting team, your primary role would be to actively aid the consulting team in different phases of the project including problem definition, effort estimation, diagnosis, solution generation and design and deployment • You will explore the alternatives to the recommended solutions based on research that includes literature surveys, information available in public domains, vendor evaluation information, etc. and build POCs • You will create requirement specifications from the business needs, define the to-be-processes and detailed functional designs based on requirements. • You will support configuring solution requirements on the products; understand if any issues, diagnose the root-cause of such issues, seek clarifications, and then identify and shortlist solution alternatives • You will also contribute to unit-level and organizational initiatives with an objective of providing high quality value adding solutions to customers. If you think you fit right in to help our clients navigate their next in their digital transformation journey, this is the place for you!\\n\\nAdditional Responsibilities:\\nAbility to work with clients to identify business challenges and contribute to client deliverables by refining, analyzing, and structuring relevant data • Awareness of latest technologies and trends • Logical thinking and problem solving skills along with an ability to collaborate • Ability to assess the current processes, identify improvement areas and suggest the technology solutions • One or two industry domain knowledge\\nTechnical and Professional Requirements:\\nPrimary skills:Cloud Platform->Microservices,Java,Java->Java API,Java->Microservices,Java-ALL->Java Cloud,Springboot\\nPreferred Skills:\\nCloud Platform->Microservices,Java->Microservices,Springboot,Java-ALL->Java Cloud,Java->Java API,Java\\nEducational Requirements\\nBachelor of Engineering,Bachelor Of Technology,Bachelor Of Science (Engg),Master Of Comp. Applications,Master Of Technology,MSc,BCA,BCom,MCOM                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                \n",
       "4  We are building the first NO-CODE AI PLATFORM for the AEC industry, with the 3rd generation Explainable AI, that allows users to create complex use cases with zero coding at the frontend and neuro-symbolic AI under the hood.\\n\\nOur NO-CODE AI PLATFORM provides end-to-end functionalities starting with specifying knowledge, to creating and annotating datasets, all the way to training and deploying ML models.\\n\\nTasks\\n\\nWe are looking for a Data Scientist who will work with our AI Modelling Team in improving the accuracy and quality of our AI models to achieve excellent performance across several use cases.\\n\\nYou will be responsible for understanding customer use cases, translating these use cases into data requirements, annotating data, training, testing and deployment of AI models.\\n\\nA substantial part of your work will involve annotating data.\\n\\nFurthermore, you will be trained to work with our No-Code AI platform to train, evaluate and deploy AI Models across several use cases - all this without writing a single line of code.\\n\\nYou will also be contributing to the improvement of these AI Models using the python/pytorch programming language.\\n\\nThe candidate will gain unique professional training and experience as a No Code AI Data Specialist.\\n\\nRequirements\\n\\nBCA / BSc. / BTech or higher with experience in data annotation and data scienceExperience in working with deep learning models using pytorch.Experience with annotation, training, testing and deployment of AI models.Ability to support several projects at one timeSelf-motivated to learn about AI pipelines and processesFluent in English writing and speakingExcellent in communicationTeam player\\n                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         \n",
       "\n",
       "   average_rating_list  company_culture  growth_opportunities  \\\n",
       "0  3.6                  3.3              3.3                    \n",
       "1  4.8                  5.0              4.0                    \n",
       "2  3.9                  3.8              3.4                    \n",
       "3  4.0                  3.8              3.7                    \n",
       "4 NaN                  NaN              NaN                     \n",
       "\n",
       "   people_you_work_with  person_you_work_for  rewards_you_receive  \\\n",
       "0  4.5                   3.5                  3.5                   \n",
       "1  5.0                   5.0                  4.0                   \n",
       "2  4.3                   4.1                  3.9                   \n",
       "3  4.4                   4.2                  3.5                   \n",
       "4 NaN                   NaN                  NaN                    \n",
       "\n",
       "   suppport_you_get  way_you_work  work_setting  \\\n",
       "0  3.0               3.5           4.0            \n",
       "1  5.0               5.0           5.0            \n",
       "2  3.8               4.0           3.9            \n",
       "3  4.0               4.2           4.0            \n",
       "4 NaN               NaN           NaN             \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           reviews  \n",
       "0  Ensure promotions are given to people who deserve it, change the loyalty gathering process, and make sure all higher ups work in teh field so they can know what works in reality and what does not.                                                                                                                                                                                                                                                                                                             \n",
       "1  Really great company who puts their employees first. Everyone is so friendly and helpful to their coworkers. The company is great at what it does, helping libraries to access the content they'd like to make available to their patrons. My colleagues were very talented and helped me to build a skillset in the IT field.                                                                                                                                                                                   \n",
       "2  Don't be so afraid of coming up with new things and seeing what the development staff can do! If we hadn't broken a few of their IT mandates in 1995, there would have been no support.dell.com worth talking about!!!! Also, don't have all of the developers in IT and crushed by budgets and IT timelines - get them out into the business units so they can be more effective and react faster to changing markets!                                                                                          \n",
       "3  I have worked for Infosys for 10 years in various roles and at various locations. The company has a wealth of resources which can be its strength when you are starting in the company. The large size of the company is a drawback for individual growth once an employee reaches middle-management in the company. I like to work in an environment where processes aid the decision. Having a large company with multiple hierarchy levels though drowns the work into processes and slow down productivity.  \n",
       "4  NaN                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                              "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "97065a77",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>average_rating_list</th>\n",
       "      <th>company_culture</th>\n",
       "      <th>growth_opportunities</th>\n",
       "      <th>people_you_work_with</th>\n",
       "      <th>person_you_work_for</th>\n",
       "      <th>rewards_you_receive</th>\n",
       "      <th>suppport_you_get</th>\n",
       "      <th>way_you_work</th>\n",
       "      <th>work_setting</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>172.000000</td>\n",
       "      <td>172.000000</td>\n",
       "      <td>172.000000</td>\n",
       "      <td>172.000000</td>\n",
       "      <td>172.000000</td>\n",
       "      <td>172.000000</td>\n",
       "      <td>172.000000</td>\n",
       "      <td>172.000000</td>\n",
       "      <td>172.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.011628</td>\n",
       "      <td>3.953488</td>\n",
       "      <td>3.526744</td>\n",
       "      <td>4.420349</td>\n",
       "      <td>4.205233</td>\n",
       "      <td>3.793023</td>\n",
       "      <td>3.987209</td>\n",
       "      <td>4.122093</td>\n",
       "      <td>4.059302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.373515</td>\n",
       "      <td>0.449743</td>\n",
       "      <td>0.500976</td>\n",
       "      <td>0.362728</td>\n",
       "      <td>0.390870</td>\n",
       "      <td>0.477516</td>\n",
       "      <td>0.488295</td>\n",
       "      <td>0.486881</td>\n",
       "      <td>0.479492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>3.975000</td>\n",
       "      <td>3.800000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>4.200000</td>\n",
       "      <td>3.600000</td>\n",
       "      <td>3.900000</td>\n",
       "      <td>4.100000</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>3.950000</td>\n",
       "      <td>3.600000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>4.200000</td>\n",
       "      <td>3.900000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.200000</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4.200000</td>\n",
       "      <td>4.300000</td>\n",
       "      <td>3.800000</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>4.300000</td>\n",
       "      <td>4.125000</td>\n",
       "      <td>4.300000</td>\n",
       "      <td>4.200000</td>\n",
       "      <td>4.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       average_rating_list  company_culture  growth_opportunities  \\\n",
       "count  172.000000           172.000000       172.000000             \n",
       "mean   4.011628             3.953488         3.526744               \n",
       "std    0.373515             0.449743         0.500976               \n",
       "min    1.000000             1.000000         1.000000               \n",
       "25%    3.975000             3.800000         3.500000               \n",
       "50%    4.000000             3.950000         3.600000               \n",
       "75%    4.200000             4.300000         3.800000               \n",
       "max    5.000000             5.000000         5.000000               \n",
       "\n",
       "       people_you_work_with  person_you_work_for  rewards_you_receive  \\\n",
       "count  172.000000            172.000000           172.000000            \n",
       "mean   4.420349              4.205233             3.793023              \n",
       "std    0.362728              0.390870             0.477516              \n",
       "min    1.000000              1.000000             1.000000              \n",
       "25%    4.400000              4.200000             3.600000              \n",
       "50%    4.400000              4.200000             3.900000              \n",
       "75%    4.500000              4.300000             4.125000              \n",
       "max    5.000000              5.000000             5.000000              \n",
       "\n",
       "       suppport_you_get  way_you_work  work_setting  \n",
       "count  172.000000        172.000000    172.000000    \n",
       "mean   3.987209          4.122093      4.059302      \n",
       "std    0.488295          0.486881      0.479492      \n",
       "min    1.000000          1.000000      1.000000      \n",
       "25%    3.900000          4.100000      4.000000      \n",
       "50%    4.000000          4.200000      4.000000      \n",
       "75%    4.300000          4.200000      4.300000      \n",
       "max    5.000000          5.000000      5.000000      "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "53067728",
   "metadata": {},
   "source": [
    "## Visualization of Null values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "afd3665c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 8))\n",
    "sns.heatmap(df.isnull(), yticklabels=False, cbar=False, cmap='cividis')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e23e1407",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Job_title               0  \n",
       "company_name            0  \n",
       "company_location        0  \n",
       "job_type                65 \n",
       "posted_date             247\n",
       "qualification           0  \n",
       "benefits                0  \n",
       "description             0  \n",
       "average_rating_list     218\n",
       "company_culture         218\n",
       "growth_opportunities    218\n",
       "people_you_work_with    218\n",
       "person_you_work_for     218\n",
       "rewards_you_receive     218\n",
       "suppport_you_get        218\n",
       "way_you_work            218\n",
       "work_setting            218\n",
       "reviews                 214\n",
       "dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "5e239bbd",
   "metadata": {},
   "outputs": [],
   "source": [
    "title = ['scien', 'Architect', 'Engineer', 'analy', 'Consul', 'Developer', 'Manager', 'Research', 'intern']\n",
    "\n",
    "dict_title = {}\n",
    "\n",
    "for i in title:\n",
    "    count = 0\n",
    "    for job_tile in df['Job_title']:\n",
    "        if i.lower() in job_tile.lower():\n",
    "            count+=1\n",
    "    dict_title[i]=count\n",
    "        \n",
    "percentage = list((np.array(list(dict_title.values()))/390) * 100)        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f131537d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIQAAAHSCAYAAACdPRB7AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAA+JUlEQVR4nO3deZgdVYE3/u/JAkOMgGBwWMUFNBBCDGF7RQTHIGRYBUFElhmWF18YHXV0FH6DrAMDgwYUELcBHSfAgLLJRDMgwiAQAwRkGQQVJIgSlKDYISTh/P64lbYDCVm6Ow2pz+d5+umqU6fuPdXn1r3V31t1qtRaAwAAAEB7DBroBgAAAACwYgmEAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGWGDHQDkuT1r3993XjjjQe6GQAAAAArjTvuuOOpWuuIRS1bYiBUSvlGkt2TPFlrHdWj/O+SHJNkfpLv1Vo/3ZR/NsnhTflHa63fX9JzbLzxxpk2bdrSbAsAAAAAS6GU8ujili3NGUIXJflSkm/2eMCdk+yVZMta65xSyjpN+WZJPphk8yTrJfnvUsqmtdb5y998AAAAAPrSEscQqrXelOT3Lyr+SJIzaq1zmjpPNuV7Jbmk1jqn1vrLJA8n2aYP2wsAAABALy3voNKbJnlXKeX2UsqPSilbN+XrJ3msR70ZTdlLlFKOKqVMK6VMmzlz5nI2AwAAAIBltbyB0JAkayXZLsmnklxWSinL8gC11q/UWsfVWseNGLHI8Y0AAAAA6AfLGwjNSPKd2jE1yQtJXp/k8SQb9qi3QVMGAAAAwCvE8gZCVybZOUlKKZsmWSXJU0muTvLBUsqqpZQ3JdkkydQ+aCcAAAAAfWRpbjs/KclOSV5fSpmR5HNJvpHkG6WUe5M8n+TQWmtNcl8p5bIk9yeZl+QYdxgDAAAAeGUpnRxnYI0bN65OmzZtoJsBAAAAsNIopdxRax23qGXLe8kYAAAAAK9SAiEAAACAlhEI9dJzzz2XbbbZJltuuWU233zzfO5zn0uS3HDDDRk7dmxGjRqVQw89NPPmzVvk+oMHD86YMWMyZsyY7Lnnnt3lBx10UEaPHp3jjjuuu+zUU0/NlVde2a/bAwAAAKz8BEK9tOqqq+aGG27I3XffnenTp2fy5Mn58Y9/nEMPPTSXXHJJ7r333rzxjW/MxRdfvMj1V1tttUyfPj3Tp0/P1VdfnSS55557stpqq+Wee+7JT37ykzzzzDN54okncvvtt2fvvfdegVsHAAAArIwEQr1USsnw4cOTJHPnzs3cuXMzePDgrLLKKtl0002TJOPHj88VV1yx1I85dOjQzJ49Oy+88EL3451wwgk56aST+mUbAAAAgHYRCPWB+fPnZ8yYMVlnnXUyfvz4bLPNNpk3b14W3Dnt8ssvz2OPPbbIdZ977rmMGzcu2223XfflYCNHjsyIESMyduzY7LHHHnn44YfzwgsvZOzYsStqkwAAAICV2JCBbsDKYPDgwZk+fXpmzZqVffbZJ/fdd18uueSSfPzjH8+cOXOyyy67ZPDgwYtc99FHH83666+fX/ziF3nPe96TLbbYIm95y1syceLE7jp77LFHLrzwwpx22mm5++67M378+Bx55JEraOsAAACAlY0zhPrQmmuumZ133jmTJ0/O9ttvn5tvvjlTp07Njjvu2H352Iutv/76SZI3v/nN2WmnnXLXXXcttPyqq67KVlttlWeffTY///nPc9lll+Xyyy9PV1dXv28PAAAAsHISCPXSzJkzM2vWrCTJ7NmzM2XKlLz97W/Pk08+mSSZM2dO/uVf/iVHH330S9Z9+umnM2fOnCTJU089lVtuuSWbbbZZ9/K5c+dm4sSJ+fSnP53Zs2enlJKkc4na888/389bBgAAAKysBEK99MQTT2TnnXfO6NGjs/XWW2f8+PHZfffdc9ZZZ2XkyJEZPXp09thjj7znPe9JkkybNi1HHHFEkuSBBx7IuHHjsuWWW2bnnXfOZz7zmYUCofPOOy+HHnpohg0bltGjR6erqytbbLFFttpqq6y55poDsbkAAADASqDUWge6DRk3blxdMAAzAAAAAL1XSrmj1jpuUctetYNKb/Wpbw50E2jccdYhA90EAAAAYBm4ZAwAAACgZQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWkYgBAAAANAyAiEAAACAlhEIAQAAALSMQAgAAACgZQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBllhgIlVK+UUp5spRy7yKWfbKUUkspr2/mSynl3FLKw6WUe0opY/uj0QAAAAAsv6U5Q+iiJLu+uLCUsmGSXZL8qkfxbkk2aX6OSnJB75sIAAAAQF9aYiBUa70pye8XsegLST6dpPYo2yvJN2vHbUnWLKWs2yctBQAAAKBPLNcYQqWUvZI8Xmu9+0WL1k/yWI/5GU3Zoh7jqFLKtFLKtJkzZy5PMwAAAABYDsscCJVShiU5LskJvXniWutXaq3jaq3jRowY0ZuHAgAAAGAZDFmOdd6S5E1J7i6lJMkGSe4spWyT5PEkG/aou0FTBgAAAMArxDKfIVRr/WmtdZ1a68a11o3TuSxsbK31N0muTnJIc7ex7ZI8U2t9om+bDAAAAEBvLM1t5ycluTXJ20opM0oph79M9euS/CLJw0m+muT/9UkrAQAAAOgzS7xkrNZ64BKWb9xjuiY5pvfNAgAAAKC/LNddxgAAAAB49RIIAQAAALSMQAgAAACgZQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWkYgBAAAANAyAiEAAACAlhEIAQAAALSMQAgAAACgZQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAts8RAqJTyjVLKk6WUe3uUnVVK+d9Syj2llO+WUtbsseyzpZSHSykPllLe10/tBgAAAGA5Lc0ZQhcl2fVFZVOSjKq1jk7ysySfTZJSymZJPphk82ad80spg/ustQAAAAD02hIDoVrrTUl+/6KyH9Ra5zWztyXZoJneK8kltdY5tdZfJnk4yTZ92F4AAAAAeqkvxhD62yT/1Uyvn+SxHstmNGUAAAAAvEL0KhAqpRyfZF6Sby/HukeVUqaVUqbNnDmzN80AAAAAYBksdyBUSjksye5JDqq11qb48SQb9qi2QVP2ErXWr9Rax9Vax40YMWJ5mwEAAADAMlquQKiUsmuSTyfZs9ba1WPR1Uk+WEpZtZTypiSbJJna+2YCAAAA0FeGLKlCKWVSkp2SvL6UMiPJ59K5q9iqSaaUUpLktlrr0bXW+0oplyW5P51LyY6ptc7vr8YDAAAAsOyWGAjVWg9cRPHXX6b+aUlO602jAAAAAOg/fXGXMQAAAABeRQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWkYgBAAAANAyAiEAAACAlhEIAQAAALSMQAgAAACgZQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWmaJgVAp5RullCdLKff2KFurlDKllPJQ8/t1TXkppZxbSnm4lHJPKWVsfzYeAAAAgGW3NGcIXZRk1xeVfSbJ9bXWTZJc38wnyW5JNml+jkpyQd80EwAAAIC+ssRAqNZ6U5Lfv6h4ryQXN9MXJ9m7R/k3a8dtSdYspazbR20FAAAAoA8s7xhCb6i1PtFM/ybJG5rp9ZM81qPejKYMAAAAgFeIXg8qXWutSeqyrldKOaqUMq2UMm3mzJm9bQYAAAAAS2l5A6HfLrgUrPn9ZFP+eJINe9TboCl7iVrrV2qt42qt40aMGLGczQAAAABgWS1vIHR1kkOb6UOTXNWj/JDmbmPbJXmmx6VlAAAAALwCDFlShVLKpCQ7JXl9KWVGks8lOSPJZaWUw5M8mmT/pvp1SSYkeThJV5K/6Yc2AwAAANALS3OXsQNrrevWWofWWjeotX691vq7Wutf1Vo3qbW+t9b6+6ZurbUeU2t9S611i1rrtP7fBADovcceeyw777xzNttss2y++eY555xzFlp+9tlnp5SSp556arGP8Yc//CEbbLBBjj322CTJnDlzsuuuu2bUqFE5//zzu+sdddRRufPOO/tnQwAAYCn0elBpAFgZDBkyJGeffXbuv//+3HbbbTnvvPNy//33J+mERT/4wQ+y0UYbvexj/NM//VN23HHH7vnvf//72WGHHXLPPffkW9/6VpLk7rvvzvz58zN27Nj+2xgAAFgCgRAAJFl33XW7Q5rXvva1GTlyZB5/vHNfhI9//OM588wzU0pZ7Pp33HFHfvvb32aXXXbpLhs6dGi6uroyd+7cdG7K2QmNTjnllH7cEgAAWDKBEAC8yCOPPJK77ror2267ba666qqsv/762XLLLRdb/4UXXsgnP/nJ/Ou//utC5ePHj88jjzyS7bbbLh/96Edz9dVXZ+zYsVlvvfX6exMAAOBlLXFQaQBok2effTb77rtvJk6cmCFDhuSf//mf84Mf/OBl1zn//PMzYcKEbLDBBguVDxkyJP/xH/+RJJk7d27e97735aqrrsonPvGJ/OpXv8ohhxySPffcs9+2BQAAFkcgBACNuXPnZt99981BBx2U97///fnpT3+aX/7yl91nB82YMSNjx47N1KlT85d/+Zfd69166625+eabc/755+fZZ5/N888/n+HDh+eMM87ornP++efnkEMOyW233ZY11lgjl156ad7znvcIhAAAGBACIQBIUmvN4YcfnpEjR+YTn/hEkmSLLbbIk08+2V1n4403zrRp0/L6179+oXW//e1vd09fdNFFmTZt2kJh0NNPP51rr7023//+93PNNddk0KBBKaVk9uzZ/bxVAACwaMYQAoAkt9xyS771rW/lhhtuyJgxYzJmzJhcd911i60/bdq0HHHEEUv12CeffHKOP/74DBo0KO973/ty8803Z4sttsjBBx/cV80HAIBlUhbc9WQgjRs3rk6bNm2Z1tnqU9/sp9awrO4465CBbgIAAADwIqWUO2qt4xa1zCVjAAyoH+347oFuAo133/SjgW4CAAAriEvGAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWkYgBAAAANAyAiEAAACAlhEIAQAAALSMQAhgMR577LHsvPPO2WyzzbL55pvnnHPOSZJ86lOfytvf/vaMHj06++yzT2bNmvWSdZ977rlss8022XLLLbP55pvnc5/7XPeygw46KKNHj85xxx3XXXbqqafmyiuv7O9NAgAASCIQAlisIUOG5Oyzz87999+f2267Leedd17uv//+jB8/Pvfee2/uueeebLrppjn99NNfsu6qq66aG264IXfffXemT5+eyZMn57bbbss999yT1VZbLffcc09+8pOf5JlnnskTTzyR22+/PXvvvfeK30gAAKCVBEIAi7Huuutm7NixSZLXvva1GTlyZB5//PHssssuGTJkSJJku+22y4wZM16ybiklw4cPT5LMnTs3c+fOTSklQ4cOzezZs/PCCy9k7ty5GTx4cE444YScdNJJK27DAACA1hMIASyFRx55JHfddVe23Xbbhcq/8Y1vZLfddlvkOvPnz8+YMWOyzjrrZPz48dl2220zcuTIjBgxImPHjs0ee+yRhx9+OC+88EJ38AQAALAiDBnoBgC80j377LPZd999M3HixKy++urd5aeddlqGDBmSgw46aJHrDR48ONOnT8+sWbOyzz775N57782oUaMyceLE7jp77LFHLrzwwpx22mm5++67M378+Bx55JH9vUkAAEDLOUMI4GXMnTs3++67bw466KC8//3v7y6/6KKLcu211+bb3/52Sikv+xhrrrlmdt5550yePHmh8quuuipbbbVVnn322fz85z/PZZddlssvvzxdXV39si0AAAALCIQAFqPWmsMPPzwjR47MJz7xie7yyZMn58wzz8zVV1+dYcOGLXLdmTNndt99bPbs2ZkyZUre/va3dy+fO3duJk6cmE9/+tOZPXt2d6g0f/78PP/88/23UQAAABEIASzWLbfckm9961u54YYbMmbMmIwZMybXXXddjj322Pzxj3/M+PHjM2bMmBx99NFJkl//+teZMGFCkuSJJ57IzjvvnNGjR2frrbfO+PHjs/vuu3c/9nnnnZdDDz00w4YNy+jRo9PV1ZUtttgiW221VdZcc82B2FwAAKBFSq11oNuQcePG1WnTpi3TOlt96pv91BqW1R1nHTLQTQBexX6047sHugk03n3Tjwa6CQAA9KFSyh211nGLWmZQaeAV551ffOdAN4HGLX93y0A3AQAA6AcuGQMAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0TK8CoVLKx0sp95VS7i2lTCql/EUp5U2llNtLKQ+XUi4tpazSV40FAAAAoPeWOxAqpayf5KNJxtVaRyUZnOSDSf4lyRdqrW9N8nSSw/uioQAAAAD0jd5eMjYkyWqllCFJhiV5Isl7klzeLL84yd69fA4AAAAA+tByB0K11seT/GuSX6UTBD2T5I4ks2qt85pqM5Ks39tGAgAAANB3enPJ2OuS7JXkTUnWS/KaJLsuw/pHlVKmlVKmzZw5c3mbAQAAAMAy6s0lY+9N8sta68xa69wk30nyziRrNpeQJckGSR5f1Mq11q/UWsfVWseNGDGiF80AAAAAYFn0JhD6VZLtSinDSiklyV8luT/JD5Ps19Q5NMlVvWsiAAAAAH2pN2MI3Z7O4NF3Jvlp81hfSfKPST5RSnk4ydpJvt4H7QQAAACgjwxZcpXFq7V+LsnnXlT8iyTb9OZxAQAAAOg/vb3tPAAAAACvMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWkYgBAAAANAyAiEAAACAlhEIAQAAALSMQAgAAACgZQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWkYgBAAAANAyAiEAAACAlhEIAQAAALSMQAgAAACgZQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALdOrQKiUsmYp5fJSyv+WUh4opWxfSlmrlDKllPJQ8/t1fdVYAAAAAHqvt2cInZNkcq317Um2TPJAks8kub7WukmS65t5AAAAAF4hljsQKqWskWTHJF9Pklrr87XWWUn2SnJxU+3iJHv3rokAAAAA9KXenCH0piQzk/xbKeWuUsrXSimvSfKGWusTTZ3fJHnDolYupRxVSplWSpk2c+bMXjQDAAAAgGXRm0BoSJKxSS6otb4jyZ/yosvDaq01SV3UyrXWr9Rax9Vax40YMaIXzQAAAABgWfQmEJqRZEat9fZm/vJ0AqLfllLWTZLm95O9ayIAAAAAfWm5A6Fa62+SPFZKeVtT9FdJ7k9ydZJDm7JDk1zVqxYCAAAA0KeG9HL9v0vy7VLKKkl+keRv0gmZLiulHJ7k0ST79/I5AAAAAOhDvQqEaq3Tk4xbxKK/6s3jAgAAANB/ejOGEAAAAACvQgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWkYgBAAAANAyAiEAAACAlhEIAQAAALSMQAgAAACgZQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWkYgBAAAANAyAiEAAACAlhEIAQAAALSMQAgAAACgZQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALdPrQKiUMriUclcp5dpm/k2llNtLKQ+XUi4tpazS+2YCAAAA0Ff64gyhjyV5oMf8vyT5Qq31rUmeTnJ4HzwHAAAAAH2kV4FQKWWDJH+d5GvNfEnyniSXN1UuTrJ3b54DAAD60jnnnJNRo0Zl8803z8SJE1+y/MYbb8waa6yRMWPGZMyYMTn55JOTJDNnzswOO+yQUaNG5corr+yuv9dee+XXv/71Cmo9APSNIb1cf2KSTyd5bTO/dpJZtdZ5zfyMJOsvasVSylFJjkqSjTbaqJfNAACAJbv33nvz1a9+NVOnTs0qq6ySXXfdNbvvvnve+ta3LlTvXe96V6699tqFyiZNmpSjjz4673//+zNhwoTsvffeueaaa/KOd7wj66233orcDADoteU+Q6iUsnuSJ2utdyzP+rXWr9Rax9Vax40YMWJ5mwEAAEvtgQceyLbbbpthw4ZlyJAhefe7353vfOc7S7Xu0KFD09XVlTlz5mTw4MGZN29eJk6cmE9/+tP93GoA6Hu9uWTsnUn2LKU8kuSSdC4VOyfJmqWUBWcebZDk8V61EAAA+sioUaNy880353e/+126urpy3XXX5bHHHntJvVtvvTVbbrlldtttt9x3331Jkg996EO56qqrMn78+Bx33HE5//zzc/DBB2fYsGErejMAoNeW+5KxWutnk3w2SUopOyX5h1rrQaWU/0yyXzoh0aFJrup9MwEAoPdGjhyZf/zHf8wuu+yS17zmNRkzZkwGDx68UJ2xY8fm0UcfzfDhw3Pddddl7733zkMPPZQ11lgj3/ve95IkTz/9dM4444x897vfzZFHHpmnn346n/zkJ7P99tsPxGYBwDLri7uMvdg/JvlEKeXhdMYU+no/PAcAACyXww8/PHfccUduuummvO51r8umm2660PLVV189w4cPT5JMmDAhc+fOzVNPPbVQnVNOOSXHH398Jk2alB122CEXX3xxTjzxxBW1CQDQa30SCNVab6y17t5M/6LWuk2t9a211g/UWuf0xXMAAEBfePLJJ5Mkv/rVr/Kd73wnH/rQhxZa/pvf/Ca11iTJ1KlT88ILL2TttdfuXv7QQw9lxowZ2WmnndLV1ZVBgwallJLZs2evuI0AgF7q7V3GAADgVWXffffN7373uwwdOjTnnXde1lxzzXz5y19Okhx99NG5/PLLc8EFF2TIkCFZbbXVcskll6SU0r3+8ccfn9NOOy1JcuCBB2bvvffOGWec0X17egB4NRAIAQDQKjfffPNLyo4++uju6WOPPTbHHnvsYte/7LLLuqfXWWed/PjHP+7bBgLACiAQAgCgX5z24f0Gugk0jv/3ywe6CQC8wvTHoNIAAAAAvIIJhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAgFeVBx98MGPGjOn+WX311TNx4sSF6px11lndy0eNGpXBgwfn97//fWbOnJkddtgho0aNypVXXtldf6+99sqvf/3rFbshA0ggBAAAALyqvO1tb8v06dMzffr03HHHHRk2bFj22Wefhep86lOf6q5z+umn593vfnfWWmutTJo0KUcffXSmTp3aHSJdc801ecc73pH11ltvALZmYAwZ6AYAAAAALK/rr78+b3nLW/LGN75xsXUmTZqUAw88MEkydOjQdHV1Zc6cORk8eHDmzZuXiRMn5pprrllRTX5FcIYQAAAA8Kp1ySWXdIc9i9LV1ZXJkydn3333TZJ86EMfylVXXZXx48fnuOOOy/nnn5+DDz44w4YNW1FNfkUQCAEAAACvSs8//3yuvvrqfOADH1hsnWuuuSbvfOc7s9ZaayVJ1lhjjXzve9/LtGnTMnbs2FxzzTXZb7/9cuSRR2a//fbLrbfeuqKaP6AEQgAAAMCr0n/9139l7NixecMb3rDYOi93BtEpp5yS448/PpMmTcoOO+yQiy++OCeeeGI/tfaVRSAEAAAAvCr1HBtoUZ555pn86Ec/yl577fWSZQ899FBmzJiRnXbaKV1dXRk0aFBKKZk9e3Z/NvkVQyAEAAAAvOr86U9/ypQpU/L+97+/u+zLX/5yvvzlL3fPf/e7380uu+yS17zmNS9Z//jjj89pp52WJDnwwANzwQUXZOutt87HPvax/m/8K4C7jAEAAACvOq95zWvyu9/9bqGyo48+eqH5ww47LIcddtgi17/sssu6p9dZZ538+Mc/7vM2vpIJhAAAAIDFuuw/txnoJtDY/wNT++yxXDIGAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAAFpGIAQAAADQMgIhAAAAgJYRCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWkYgBAAAANAyAiEAAACAlhEIAQAAALSMQAgAAACgZQRCAAAAAC0jEAIAAABomeUOhEopG5ZSflhKub+Ucl8p5WNN+VqllCmllIea36/ru+YCAPSP+fPn5x3veEd23333RS6/7LLLstlmm2XzzTfPhz70oSTJgw8+mK222iqjR4/OrbfemiSZN29e3vve96arq2uFtR0AYFkN6cW685J8stZ6ZynltUnuKKVMSXJYkutrrWeUUj6T5DNJ/rH3TQUA6D/nnHNORo4cmT/84Q8vWfbQQw/l9NNPzy233JLXve51efLJJ5MkF154Yc4555xsvPHG+djHPpYrrrgiF1xwQT784Q9n2LBhK3oTAACW2nKfIVRrfaLWemcz/cckDyRZP8leSS5uql2cZO9ethEAoF/NmDEj3/ve93LEEUcscvlXv/rVHHPMMXnd6zonPq+zzjpJkqFDh6arqytdXV0ZOnRoZs2alWuuuSaHHHLICms7AMDy6M0ZQt1KKRsneUeS25O8odb6RLPoN0ne0BfPAQDQX/7+7/8+Z555Zv74xz8ucvnPfvazJMk73/nOzJ8/PyeeeGJ23XXXHHPMMTnkkEMyZ86cXHjhhTnllFNy3HHHZdAgwzQCAK9svT5aKaUMT3JFkr+vtS50jnWttSapi1nvqFLKtFLKtJkzZ/a2GQAAy+Xaa6/NOuusk6222mqxdebNm5eHHnooN954YyZNmpQjjzwys2bNykYbbZQbb7wxt956a4YNG5YZM2Zk5MiROfjgg3PAAQd0B0kAAK80vQqESilD0wmDvl1r/U5T/NtSyrrN8nWTPLmodWutX6m1jqu1jhsxYkRvmgEAsNxuueWWXH311dl4443zwQ9+MDfccEM+/OEPL1Rngw02yJ577pmhQ4fmTW96UzbddNM89NBDC9U5/vjjc+qpp+bcc8/NEUcckTPPPDMnnXTSitwUAICl1pu7jJUkX0/yQK318z0WXZ3k0Gb60CRXLX/zAAD61+mnn54ZM2bkkUceySWXXJL3vOc9+fd///eF6uy999658cYbkyRPPfVUfvazn+XNb35z9/If/ehHWW+99bLJJpukq6srgwYNyqBBg9xpDAB4xerNGELvTHJwkp+WUqY3ZcclOSPJZaWUw5M8mmT/XrUQAGAAnHDCCRk3blz23HPPvO9978sPfvCDbLbZZhk8eHDOOuusrL322kmSWmtOPfXUXHrppUmSo446KgcddFDmzZuXCy64YCA3AQBgsZY7EKq1/k+SspjFf7W8jwsAMFB22mmn7LTTTkmSk08+ubu8lJLPf/7z+fznP/+SdUopmTJlSvf8yJEjc+edd/Z7W4Ele+6557Ljjjtmzpw5mTdvXvbbb7+XXMr5q1/9KoceemhmzZqV+fPn54wzzsiECRNyyy235CMf+UhWWWWVTJo0KZtssklmzZqV/fffP5MnTzZ4PPCq1yd3GQMAWBpf+uQ1A90EGseevcdANwH63aqrrpobbrghw4cPz9y5c7PDDjtkt912y3bbbddd59RTT83++++fj3zkI7n//vszYcKEPPLIIzn77LNz3XXX5ZFHHsmXv/zlnH322Tn11FPdSRBYaXgnAwAAVkqllAwfPjxJMnfu3MydOzedoVAXrvOHP3RulvzMM89kvfXWS5IMHTo0XV1d6erqytChQ/Pzn/88jz32WPdZhACvds4QAgAAVlrz58/PVlttlYcffjjHHHNMtt1224WWn3jiidlll13yxS9+MX/605/y3//930mSz372sznkkEOy2mqr5Vvf+lb+4R/+IaeeeupAbAJAv3CGEAAAsNIaPHhwpk+fnhkzZmTq1Km59957F1o+adKkHHbYYZkxY0auu+66HHzwwXnhhRcyZsyY3HbbbfnhD3+YX/ziF1l33XVTa80BBxyQD3/4w/ntb387QFsE0DcEQgAAwEpvzTXXzM4775zJkycvVP71r389++/fuTHy9ttvn+eeey5PPfVU9/IFdxL8p3/6p5x00kk588wzc+SRR+bcc89doe0H6GsCIQAAYKU0c+bMzJo1K0kye/bsTJkyJW9/+9sXqrPRRhvl+uuvT5I88MADee655zJixIju5d/85jczYcKErLXWWunq6sqgQYMyaNCgdHV1rbDtAOgPxhACAABWSk888UQOPfTQzJ8/Py+88EL233//7L777jnhhBMybty47Lnnnjn77LNz5JFH5gtf+EJKKbnooou6B57u6urKRRddlB/84AdJkk984hOZMGFCVllllfzHf/zHQG4aQK8JhAAAgJXS6NGjc9ddd72k/OSTT+6e3myzzXLLLbcscv1hw4blhz/8Yff8u971rvz0pz/t+4YCDACBEAAA0GsPnHbDQDeBxsjj3zPQTQBeBYwhBAAAANAyAiEAAACAlhEIAQAAALSMQAgAAACgZQRCAAAAAC0jEAIAAABoGYEQAAAAQMsIhAAAAABaRiAEAAAA0DICIQAAAICWEQgBAAAAtIxACAAAAKBlBEIAAAAALSMQAgAAAGgZgRAAAABAywiEAAAAWCn87d/+bdZZZ52MGjVqkctvvPHGrLHGGhkzZkzGjBmTk08+OUkyc+bM7LDDDhk1alSuvPLK7vp77bVXfv3rX6+IpsMKJxACAABgpXDYYYdl8uTJL1vnXe96V6ZPn57p06fnhBNOSJJMmjQpRx99dKZOnZqJEycmSa655pq84x3vyHrrrdffzYYBMWSgGwAAAAB9Yccdd8wjjzyyzOsNHTo0XV1dmTNnTgYPHpx58+Zl4sSJueaaa/q+kfAK4QwhAAAAWuPWW2/Nlltumd122y333XdfkuRDH/pQrrrqqowfPz7HHXdczj///Bx88MEZNmzYALcW+o8zhAAAAGiFsWPH5tFHH83w4cNz3XXXZe+9985DDz2UNdZYI9/73veSJE8//XTOOOOMfPe7382RRx6Zp59+Op/85Cez/fbbD3DroW85QwgAAIBWWH311TN8+PAkyYQJEzJ37tw89dRTC9U55ZRTcvzxx2fSpEnZYYcdcvHFF+fEE08cgNZC/xIIAQAA0Aq/+c1vUmtNkkydOjUvvPBC1l577e7lDz30UGbMmJGddtopXV1dGTRoUEopmT179kA1GfqNS8YAAABYKRx44IG58cYb89RTT2WDDTbISSedlLlz5yZJjj766Fx++eW54IILMmTIkKy22mq55JJLUkrpXv/444/Paaed1v1Ye++9d84444zu29PDykQgBAAAwEph0qRJL7v82GOPzbHHHrvY5Zdddln39DrrrJMf//jHfdY2eKURCAEAALBMjKnzyqEvWF7GEIJemDx5ct72trflrW99a84444yXLL/ooosyYsSIjBkzJmPGjMnXvva1JMmDDz6YrbbaKqNHj86tt96aJJk3b17e+973pqura4VuAwAAAO3jDCFYTvPnz88xxxyTKVOmZIMNNsjWW2+dPffcM5ttttlC9Q444IB86UtfWqjswgsvzDnnnJONN944H/vYx3LFFVfkggsuyIc//OEMGzZsRW4GAAAALeQMIVhOU6dOzVvf+ta8+c1vziqrrJIPfvCDueqqq5Zq3aFDh6arqytdXV0ZOnRoZs2alWuuuSaHHHJIP7caAAAAnCEEy+3xxx/Phhtu2D2/wQYb5Pbbb39JvSuuuCI33XRTNt1003zhC1/IhhtumGOOOSaHHHJI5syZkwsvvDCnnHJKjjvuuAwaJKMFAACg//nvE/rRHnvskUceeST33HNPxo8fn0MPPTRJstFGG+XGG2/MrbfemmHDhmXGjBkZOXJkDj744BxwwAH52c9+NsAtBwAAYGUmEILltP766+exxx7rnp8xY0bWX3/9heqsvfbaWXXVVZMkRxxxRO64446XPM7xxx+fU089Neeee26OOOKInHnmmTnppJP6t/EAAAC0mkAIltPWW2+dhx56KL/85S/z/PPP55JLLsmee+65UJ0nnniie/rqq6/OyJEjF1r+ox/9KOutt1422WSTdHV1ZdCgQRk0aJA7jQEAANCvjCEEy2nIkCH50pe+lPe9732ZP39+/vZv/zabb755TjjhhIwbNy577rlnzj333Fx99dUZMmRI1lprrVx00UXd69dac+qpp+bSSy9Nkhx11FE56KCDMm/evFxwwQUDtFUAAAC0gUAIemHChAmZMGHCQmUnn3xy9/Tpp5+e008/fZHrllIyZcqU7vmRI0fmzjvv7J+GAgAAQA8CIV4VfnXyFgPdBBobnfDTgW4CAAAAvWQMIQAAAICW6bdAqJSyaynlwVLKw6WUz/TX8wAAAACwbPolECqlDE5yXpLdkmyW5MBSymb98VwAAAAALJv+OkNomyQP11p/UWt9PsklSfbqp+cCAAAAYBn0VyC0fpLHeszPaMoAAAAAGGCl1tr3D1rKfkl2rbUe0cwfnGTbWuuxPeocleSoZvZtSR7s84a88r0+yVMD3QhWGP3dLvq7XfR3u+jvdtHf7aK/20V/t0tb+/uNtdYRi1rQX7edfzzJhj3mN2jKutVav5LkK/30/K8KpZRptdZxA90OVgz93S76u130d7vo73bR3+2iv9tFf7eL/n6p/rpk7CdJNimlvKmUskqSDya5up+eCwAAAIBl0C9nCNVa55VSjk3y/SSDk3yj1npffzwXAAAAAMumvy4ZS631uiTX9dfjryRafclcC+nvdtHf7aK/20V/t4v+bhf93S76u13094v0y6DSAAAAALxy9dcYQgAAAAC8QgmElkEp5fhSyn2llHtKKdNLKdsupt64Usq5vXie4140/+NlqU/fKKXsXUqppZS3L+N6j5RSXr+I8j1LKZ/p8dibLWe71iyl/L/lWZdFK6XMb/bpBT+f6cVjvez+ysqrlHJYKeVLy7HeTqWU/9MfbWqjUspfllIuKaX8vJRyRynlulLKpv38nBuXUu5tpseUUib08vF8ri+DHu/h95VS7i6lfLKU0qfHuD37mFeO5jjt33vMDymlzCylXDuQ7aLv9djP7y2lXFNKWXOg27RAKeXGUoo7V61AS3O8XUr5+1LKsBXRnlczgdBSKqVsn2T3JGNrraOTvDfJY4uqW2udVmv9aC+ebqEDwVrrkv5RcODYPw5M8j/N74WUUpZ5/K1a69W11jOa2b2TLFcglGTNJAKhvjW71jqmx88ZS15l0ZZif+2V5Xnt8Yq3UxKBUB8opZQk301yY631LbXWrZJ8NskbVmAzxiTpVSAUn+vLasF7+OZJxifZLcnnBrhNy6yUMnig2/Aq9Kcko0opqzXz45M8PoDtWSSf3X1iwX4+KsnvkxyzIp+8dPjf+RViKY+3/z7JMgVCbXwf9qJeeusmearWOidJaq1P1Vp/XUrZupTy4+YbqamllNc23/ZemySllNeUUr7RLLurlLJXU35YKeU7pZTJpZSHSilnNuVnJFmtScC/3ZQ92/xet5RyU490/F2Lqk/vlVKGJ9khyeFJPtiU7VRKubmUcnWS+0spg0sp/9r0xT2llL/r8RB/V0q5s5Ty0wVnGC04e6A5E2DPJGc1/faW5mdy8232zT3WeUMp5bvN6+vuZt0zkrylWfesFfhnaZ3mbK+TFtGXI0opU5pvo79WSnm0NGeF9dhfd2q+Mbq8lPK/pZRvN/+sppSyVSnlR01/f7+Usm5TvrjXwUWllC+XUm5PcuaA/DFWcqWUK5u/+32llKOasmdLKac1+95tpZQ3NOV7lFJub97T/3tBeY/Hem0p5ZellKHN/OoL5kspHy2l3N+8Z1xSStk4ydFJPt7s0+9awZu+stk5ydxa65cXFNRa707yP6WUs5r365+WUg5IlrifntGjr/61KbuolLLfgsdesL/3mF8lyclJDmj684BSyjallFub18uPSylva+ou9XEAS6/W+mSSo5IcWzoGN33/k6Yv/2+SNPvfXy9Yb0HfLq5+T6WUvyil/FvzWrqrlLJzU35YKeWq5jX1UCnlcz3W+XDpHAtOL6VcWJp/Opr3mbNLKXcn2b6f/zwrq+uSLOjLA5NMWrBgWfe/ZtkFpZRpzefBST3KJzTvE3eUUs4tS3esf3Up5YYk1/f/n6FVbk2yfvKyx04faN7z7y6l3NSULe79YHgp5fry5+O9BX24cSnlwVLKN5Pcm2TDUso/NnXubt6vF/hA8xr4WfFZ3u/KEo63SykfTbJekh+WUn7Y1N2leT+4s5Tyn6Xz/96C4/1/KaXcmU4/LvL4f6VVa/WzFD9JhieZnuRnSc5P8u4kqyT5RZKtmzqrp3Pntp2SXNuU/XOSDzfTazbrvybJYc26ayT5iySPJtmwqffsi5772eb3J5Mc30wPTvLaRdX30yf9fVCSrzfTP06yVdOvf0rypqb8I0kuTzKkmV+r+f1Ikr9rpv9fkq8104cl+VIzfVGS/Xo83/VJNmmmt01yQzN9aZK/79HnayTZOMm9A/03Wpl+ksxv9u8FPwcsoS+/lOSzzfSuSWqS1zfzC/bXnZI8k2SDdML3W9MJGYc2r6kRTb0DknxjCa+Di5Jcm2TwQP+tVtafHvvvaukc9K3d9OseTfmZSf6/Zvp1+fNNGY5IcnYz3XMf/7ckezfTR/Wo8+skqzbTaza/T0zyDwP9N1gZfpJ8NMkXFlG+b5IpzfvoG5L8Kp0veha3n66d5MEe/bygr1783r1gf+9+X+75OmjmV8+fPyfem+SKHvWW6jjAzxL7/SV/rySzmr4+qse+u2qSaUnelGSfJBc35aukc9b3ai9Tv2cffzJ/ft9+e/N6+oumT59oXj8L3kvGJRmZ5JokQ5t1zk9ySDNdk+w/0H/DV+tPkmeTjE7neOwv0vkM3yl/Pg5fnv1vwefB4CQ3No//F81rZMEx4KQs3bH+jAWP56f3fd2jX/4zya7N/OKOnX6aZP0F/dL8Xtz+PSTJ6k3565M8nKQ0+/0LSbZrlu2WzjHcsBe9Vm7Mnz/nJyT574H+e63sP1nC8Xaz7JH8+fj89UluSvKaZv4fk5zQo96nezz2I1nE8f/K+uP0xaVUa322lLJVknel8w3kpUlOS/JErfUnTZ0/JEnpfLm4wC5J9iyl/EMz/xdJNmqmr6+1PtOsc3+SN2Yxl6E1fpLkG6XzrfOVtdbpfbBpLNqBSc5ppi9p5q9NMrXW+sum/L1JvlxrnZcktdbf91j/O83vO5K8/+WeqEmn/0+S/+zx2lm1+f2eJIc0jz8/yTOllNct5zaxeLNrrWMWs2xRfblDOv9MpNY6uZTy9GLWnVprnZEkpZTp6RxYzEoyKsmUpr8HJ3liCa+DJPnP5jVA//hoKWWfZnrDJJskeT6d/T7p9P/4ZnqDJJeWzpldqyT5ZV7qa0k+neTKJH+T5Mim/J4k3y6lXNksY8XYIcmkZh/6bSnlR0m2TvKHLHo/vS3Jc0m+3pwF0JvxSNZIcnEpZZN0/vkf2mPZsh4HsOx2STK6/PnMrjXS2b//K8k5pZRV0wn2b6q1zi6lLK7+z3o85g5Jvpgktdb/LaU8mmTBOFVTaq2/S5JSyneauvPS+WLpJ837+2pJnmzqz09yRd9ucrvUWu8pnbMtD0znbKGelmf/2790zhQdkk5wvFk6/2j+oscx4KR0woXk5Y/1p7zo+JDlt1rzHr1+kgfSOY56uWOnW5JcVEq5LH8+llvc/j0jyT+XUnZMJwBaP3++1PjRWuttzfR7k/xbrbUredlj/417vbUsi0V9jv/Pi+psl86+fEvzWlklnfBogUtfVH+p/5d7tRMILYPmQPLGJDeWUn6apbt2tSTZt9b64EKFnQGp5/Qomp8l9Eet9abmjeqv03mD+3yt9ZvLsAkshVLKWukEMVuUUmo6/7DXJN9L5wyhpbGgb5fYr+kcZMx6mUCCgbUsfbm4dXuuX5LcV2td6NKAUsrqefnXwdK+9lhGpZSd0jnI277W2lVKuTGdA/q5tfl6KAv3/xeTfL7WenWz7okvfsxa6y3NqeY7pXNm14LBaP86yY5J9khyfClli77fola7L8l+S6y1sJfsp7XWeaWUbZL8VfN4x6bzuTAvzeX2pTOWxCpL8finJPlhrXWf5p/WG1/uuZex7SxCKeXN6fw9n0znPffvaq3fX0S9G5O8L50zNS9ZULyo+k3fLY26iPmSztlIn11E/eeE/X3i6iT/ms7ZAmv3KF+m/a+U8qYk/5DO2f9Pl1IuSufz4OW83LG+z+6+M7vWOqZ0Bgn+fjr/h12UxRw71VqPbvrgr5Pc0Xyxv7j9+7AkI5JsVWudW0p5JH/u9/449qdvLc1naUknoH3J2LCNF/dza/rTGEJLqZTytubbhQXGpJNOr1tK2bqp89ry0kHjvp/OeDILxiR4x1I83dzmLKAXt+GNSX5ba/1qOt8+j325+iy3/ZJ8q9b6xlrrxrXWDdM5A+DF1wNPSfJ/F/R5EyQtrT8meW3SfWbZL0spH2gep5RStmzqXZ/OpWkLrnteo+e6DJhbkuyfdK5HTucSoqX1YJIRpTNQfUpnXJnNl/A6oH+tkeTpJgx6ezrfIi2p/oJBSw99mXrfTPIf6Vw+tiBA2LDW+sN0TlVeI53Lke3TfeeGJKs23+4nSUopo9M5M++A5n10RDqh3NTFPUjzrfMatdbrknw8yYJ98ZF0zvRIOmPBLeqz98X92fP1cthSbofP9eXU9O+X07lsr6ZzHPaR8ucxvTYtpbymqX5pOmfwvSvJ5Kbs5eovcHM6l5andO5gt1E67+1JMr6UslbpDHK8dzqfF9cn2a+Usk6zzlrNMR195xtJTqq1/vRF5cu6/62ezj+Gz5TO+HC7NeUPJnlzj2DwgB7rLM+xPsupOTvno+lcutmVxRw7lVLeUmu9vdZ6QpKZ6Zz9u7j9e40kTzZh0M7pnC22KFOS/E0TSi3rsT8rXs/P49uSvLOU8take+yvfr0D6auFQGjpDU/nlNP7Syn3pHPK2QnpfCB8sXQGA5ySl36LcEo6B4z3lFLua+aX5CtN/RcPJrlTkrtLKXc1z3vOEuqzfA5M5y41PV2Rl95t7GvpjBtwT9P/H1qG57gkyadKZ/DBt6RzYHl48zj3JdmrqfexJDs3Z6TdkWSz5lT0W0pnoDyDSveNBQO4LvhZ0l3GTkqyS+ncgvgDSX6TzofOEtVan08ndPyXpr+n5893mFrc64D+NTmdb4YfSGfQ9tuWUP/EdE5PvyPJUy9T79vphIULBjgdnOTfm/35riTn1lpnpTO2yD7FoNK91gQA+yR5b+ncdv6+JKenE8zdk+TudEKjT9daf/MyD/XaJNc2n/f/k+QTTflXk7y7/HkA4EV9c/zDJJs1/XlAOuNPnd58di/tt4w+15fNgvfw+5L8d5IfpPM+nXQ+q+9Pcmfznn1h/twPP0hnTMj/bt6bl1R/gfOTDGr25UuTHFabm46kEzRekc7r7YraufPs/Un+vyQ/aF5TU9K5FIk+UmudUWs9dxGLlmn/q51B6O9K8r/pvG/c0pTPTmcskcnNe/8f0xm3JFm+Y316odZ6Vzr72IFZ/LHTWaUzIPC96Yz7c3cWv39/O8m4Zp8+JJ3+X9TzTk7nbLRppXNp0j8sqh6vGF9JZ5/9Ya11Zjqh8KTmffjWdMaAa70FgyUCsJRKZ8yJ+c1lJdsnucAlf7xY6YxRsFet9eCBbgvQ/0rnspNxtdZjB7ot9L1SyvBmTNGS5LwkD9VavzDQ7QLojZX6ejiAfrJRksuaS4Cez58HDIYkSSnli+lcajBhoNsCQJ84spRyaDpjh92VztklAK9qzhACAAAAaBljCAEAAAC0jEAIAAAAoGUEQgAAAAAtIxACAAAAaBmBEAAAAEDLCIQAAAAAWub/B1Q3+jzYcJ0GAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "label = ['Scientist', 'Architect', 'Engineer', 'analyst', 'Consultant', 'Developer', 'Manager', 'Research', 'intern']\n",
    "value = list(dict_title.values())\n",
    "plt.figure(figsize=(20, 8))\n",
    "#sns.barplot(x=label, y=value)\n",
    "\n",
    "ax = sns.barplot(x=label, y=value, palette='tab10')\n",
    "patches = ax.patches\n",
    "for i in range(len(patches)):\n",
    "   x = patches[i].get_x() + patches[i].get_width()/2\n",
    "   y = patches[i].get_height()+0.5\n",
    "   ax.annotate('{:.1f}%'.format(percentage[i]), (x, y), ha='center')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1eda79d",
   "metadata": {},
   "source": [
    "Jobs with Scientist label are predominant amont all other jobs. Nearly 39.5% jobs are in scientist/science domain since it is the booming domain and Data Scientists are paid the highest among all other designations. Followed by Scientists, Analysts/Analysis job occurs. 24% jobs are in Analyst roles since Analysts are second highest paid followed by Data scientists. Architect domain is least of 0.5% "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d8292afe",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['company_location']=df['company_location'].replace(\"India\", \"India, India\")\n",
    "df['company_location']=df['company_location'].replace(\"Remote\", \"Remote, Remote\")\n",
    "df['company_location']=df['company_location'].replace(\"Maharashtra\", \"Maharashtra, Maharashtra\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01cbc3f1",
   "metadata": {},
   "source": [
    "## Feature Engineering"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82a511f7",
   "metadata": {},
   "source": [
    "### Feature Extraction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "cf8443e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.insert(3, \"City\", df['company_location'].apply(lambda x :x.split(',')[-2:][0]))\n",
    "df.insert(4, \"State\", df['company_location'].apply(lambda x :x.split(',')[-2:][1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4e5ae97",
   "metadata": {},
   "source": [
    "City and state are extracted from company_location"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "460d1d60",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[df['Job_title'].map(lambda x : 'scien' in x.lower()), 'Designation']=\"Scientist\"\n",
    "df.loc[df['Job_title'].map(lambda x : 'archi' in x.lower()), 'Designation']=\"Architect\"\n",
    "df.loc[df['Job_title'].map(lambda x : 'engineer' in x.lower()), 'Designation']=\"Engineer\"\n",
    "df.loc[df['Job_title'].map(lambda x : 'analy' in x.lower()), 'Designation']=\"Analyst\"\n",
    "df.loc[df['Job_title'].map(lambda x : 'consul' in x.lower()), 'Designation']=\"Consultant\"\n",
    "df.loc[df['Job_title'].map(lambda x : 'develop' in x.lower()), 'Designation']=\"Developer\"\n",
    "df.loc[df['Job_title'].map(lambda x : 'manager' in x.lower()), 'Designation']=\"Manager\"\n",
    "df.loc[df['Job_title'].map(lambda x : 'research' in x.lower()), 'Designation']=\"Research\"\n",
    "df.loc[df['Job_title'].map(lambda x : 'intern' in x.lower()), 'Designation']=\"Intern\"\n",
    "df.loc[df['Job_title'].map(lambda x : 'admin' in x.lower()), 'Designation']=\"Administrator\"\n",
    "df.loc[df['Job_title'].map(lambda x : 'associat' in x.lower()), 'Designation']=\"Associate\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4cc4d41c",
   "metadata": {},
   "source": [
    "Designations are extracted from Job Titles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "e953aa5d",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Designation'] = df['Designation'].fillna('Other')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4ae9fa83",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "percentage = list(np.round((df['State'].value_counts().values)/390 * 100, 2))\n",
    "\n",
    "plt.figure(figsize=(20, 10))\n",
    "ax = sns.barplot(y=df['State'].value_counts().values, x=df['State'].value_counts().index)\n",
    "patches = ax.patches\n",
    "for i in range(len(patches)):\n",
    "   x = patches[i].get_x() + patches[i].get_width()/2\n",
    "   y = patches[i].get_height()+0.5\n",
    "   ax.annotate('{:.1f}%'.format(percentage[i]), (x, y), ha='center')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78e3a0c2",
   "metadata": {},
   "source": [
    "From the visualization we can infer that Data Science jobs are found more in Karnataka. 51% Jobs are posted from Karnataka. Orissa, Goa & Andhra pradesh has 0.3% of job postings"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "44e7cdf2",
   "metadata": {},
   "source": [
    "Since Karnataka has higher job posting we can do a deep analayis with respect to Karnataka"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "3baa98e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['State']=df['State'].apply(lambda x : x.strip())\n",
    "df['City']=df['City'].apply(lambda x:x.strip())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "5b21b115",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[df['State']=='Karnataka']['City'].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ce363821",
   "metadata": {},
   "source": [
    "In karnataka, Bengaluru and Mysore has job posting in which more job postings are in Bangalore since it is the electronic city"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "26b62099",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 8))\n",
    "label = df[df['City']=='Bengaluru']['Designation'].value_counts().index\n",
    "values = df[df['City']=='Bengaluru']['Designation'].value_counts().values\n",
    "plt.title('Designations of Job in Bangalore')\n",
    "plt.pie(x=values, labels=label, autopct='%.2f');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "de2800d9",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['job_type'] = df['job_type'].fillna(\"Not mentioned\")\n",
    "df['posted_date'] = df['posted_date'].fillna(\"Not mentioned\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "e7acbbf6",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['description'] = df['description'].apply(lambda x:x.replace('\\n', ' '))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "632d0947",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['description'] = df['description'].apply(lambda x:' '.join(re.sub(\"([^0-9A-Za-z \\t])|(\\w+:\\/\\/\\S+)\",\" \",x).split()))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "54959fab",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['description'] = df['description'].str.lower()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "fb682113",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.insert(10, 'desc_tokenized', df['description'].apply(lambda x:word_tokenize(x)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "9b87dc2f",
   "metadata": {},
   "outputs": [],
   "source": [
    "stop = stopwords.words('english')\n",
    "df.insert(11, 'no_stopwords', df['desc_tokenized'].apply(lambda x:' '.join([word for word in x if word not in stop])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "beb53a3f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "db67a6ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "porter_stemmer = PorterStemmer()\n",
    "def stemming(text):\n",
    "    stem_text=[porter_stemmer.stem(word) for word in text.split()]\n",
    "    return stem_text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "15c3352a",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.insert(12, 'desc_stemmed', df['no_stopwords'].apply(lambda x: stemming(x)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "6b11a31a",
   "metadata": {},
   "outputs": [],
   "source": [
    "wordnet_lemmatizer = WordNetLemmatizer()\n",
    "def lemmatizer(text):\n",
    "    lemm_text = [wordnet_lemmatizer.lemmatize(word) for word in text.split()]\n",
    "    return lemm_text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "47c1b3ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.insert(13, 'desc_lemmatized', df['no_stopwords'].apply(lambda x:lemmatizer(x)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "ffdabd16",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.insert(14, 'most_common_words', df['desc_lemmatized'].apply(lambda x : collections.Counter(x).most_common(1)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "fca4f18e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh0AAAD1CAYAAAD06EchAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAa/klEQVR4nO3debgkdX3v8fcHBkQWceGoDDCOGsUo7qMJoMJ1xZ0kGjUuTFDR5MYtblFJNIlJNJpoIvF6UREVBBXFGE2ixCsOCi4DgYACgqAM+yAqICog3/tH/Y42Tc85M6d76sz0vF/P08+p5ddV3676dfenq6r7pKqQJEna2LZa7AIkSdKWwdAhSZJ6YeiQJEm9MHRIkqReGDokSVIvDB2SJKkXhg5tVEnekuSoxa5jsSU5McmLFruOdZmrviRvTPKBnupYmeSrC7zvI5OcO+m2W5okz03yxUVad299TYvD0LGFSfKGJP8xNO28dUx7dr/VaVNUVX9bVWMFphY+K8lvTaquYVV1UlXtOcm2kwjNSfZPcvE4yxixzJVJfpnkuna7MMmHktx73GVX1dFV9fhJ1DmXUdtlEn1NmzZDx5ZnFbBPkq0BkuwKbAM8eGjab7S26y3JkgnXqimQJMALgKvb36mRzkZ9HZ3jeXVKVe0I7Aw8FvgZcGqSvTZmPdI4DB1bnm/RhYwHtfFHAl8Gzh2a9r2qujTJ0iSfTXJ1kvOTvHh2Qe1T4HFJjkpyDbAyyd2TfCXJtUlOAHYZaL9da/vDJD9O8q0kdxlVZJI9knw6ydrW/rA2faskhyb5QZIrk3wkyc5t3vL2afoPk6xJ8qMkL03ysCT/09Z52MA6Vib5WpJ3tXkXJNmnTV/Tln/QQPud2/rWtvUfOvuGM3taIMk723ovTPLEEY9r27Yt7z8w7c5Jrk8yM6L9LT5pDzzGJQPrvaBt7wuTPHeg7cFJzm71fCHJ3QbmPS7JOUl+0rZJRu2H4RoG1n9QkouSXJXkTeu6b/NIYFfg5cCzk2w7sOw7tf51TZJvAvccWncl+eN0R96uTfLXSe6Z5OR2n0/MLi9Dn5yTfD/Ja9q+/0mSjyfZbh1tX5/kkraOc5M8JskBwBuBZ6U7mnBGa3tikr9J8jXgeuAerc+d3e5/QZKXtLY7AP8BLM2vj0osTXKbJO9Ocmm7vTvJbQZrazVdDnxoro1bVb+squ9V1R8DXwHeMvC4frttqx8nOSPJ/gPzRvadDJ3iSvL4tk1+kuS96Z7fLxpsu65+v4DtMtzfn5bk263+E5P85vrsX23CqsrbFnajCxmvasOHAQcDfzM07Yg2vAp4L7AdXShZCzy6zXsLcCNwIF2AvS1wCvCPwG2ARwHXAke19i8B/g3YHtgaeChwuxH1bQ2cAbwL2KGt+xFt3sHA+cA9gB2BTwMfbfOWAwW8r93n8cDPgc8AdwZ2A64E9mvtVwI3AX/Y1vlW4CLgX1r9j2/179jafwT4V2Cntq7vAi8cWNaNwIvbsv4IuBRIm38i8KI2/F7g7QOP9xXAv61jX71ldvsNPcYlbdtcA+zZ5u0K3K8NP71tp99sbQ8FTm7zdmmP6xl0AfRVbTu8aL4aBtb//ra/Hwj8AvjNOfrbB4FPtHX9EPi9gXnHtnk7AHsBlwBfHZhfbZvfDrhfW9eX2v7fGfgOcFBruz9w8cB9vw98E1gK3BE4G3jpcFtgT2ANsHTgMd5z1PYf2JcXtXqWtMf1ZLrAFGA/ujDykFF1tWl/BXydrl/OACcDfz3Q/ibg7XT98LYjtunKwe00MP1g4Io2vFvb3k+ie34+ro3PMHff+dWy6frKNcDvtsf6Crp+/qKBtnP1+w3dLr/a3sC9gZ+2urcBXkfXp7edb/9623Rvi16At0XY6d0T+/g2fAZwL+CAoWkHAXsAvwR2Grjv3wFHDixn1cC8Ze3FcoeBaR8beBE5uL24PmCe+vamCzdLRsz7EvDHA+N7the9Jfz6DXG3gfk/BJ41MP4p4JVteCVw3sC8+7f732Xo/g9qL6g3APcdmPcS4MSBZZ0/MG/7tqy7tvETB16of4vuTWv2hXk18Ptz7Ku5QsePgd9j6I2J7lPkCwfGt6J7wb8b3SmOrw/MC3AxGxY6dh+Y/03g2eu47/Z0b1oHtvH/C/xrG9667bv7DLT/W24dOvYdGD8VeP3A+D8A727D+3Pr0PG8gfG/B9433JbuVOKVdKcotplr+w/sy7+apw9/BnjFqLratO8BTxoYfwLw/YH2NwDbzbH8lYwOHQcAN7bh19MC+cD8L9A9t+fqO79adusrpwz1lTXcMnSss98vYLsM9rU/Bz4x1IcvAfafb/9623Rvnl7ZMq0CHpHkjsBMVZ1HFwb2adP2am2WAldX1bUD9/0B3SeoWWsGhpcCP6qqnw61n/VRuhe9Y9sh5b9Pss2I+vYAflBVN42Yt3RomT+gewMePE1zxcDwz0aM7zhHW6pqVPtd6D5tDa97cFtcPjtQVde3wcF1zc77Bl0A2D/Jfeje9D473G4+bTs/C3gpcFmSz7flQRcu/qkdlv4x3fUUafUuZWC/VfeKPbgf18flA8PXM+JxNr9DF0T/vY0fDTwx3amkGbp9N7juH3BrG7I/N7jOqjofeCXdG96VSY5NsnSOZcLQ9kryxCRfT3fq7Md0Rxd2GXnPzqh+PLjOtVX183lqGGU3un0NXR945mwfaHU9Ath1nr4zXOdwXxm+KHad/X4B22V43b/aRlV1c6tl5HOOufuhNhGGji3TKXSHpl8MfA2gqq6hOyz6YuDSqrqwjd8xyU4D911G92ljVg0MXwbcoZ2vHWxPW8eNVfWXVXVfYB/gKYy+sHANsCyjL6C7lO7FdHD5N3HLN6KN4Sq6T+XD675kdPN5fRh4HvB84Lg53mB+SvfpcdZdB2dW1Req6nF0h8fPoTvtAd02fElV3X7gdtuqOpluP+0xu4wkGRyfsIPo3gguatcnfJIuvP0B3dGsm4bWvexWS+hBVX2sqh5Bt3+L7tQG3LJ/3+IuswPtWoxPAe+kO0p2e7qQleG2A0b140tHLX8D/Q5wUhteQ3ekY7AP7FBVb4M5+86gy4DdZ0daX9l9RLtbWeB2GXSLbTTQTxf6nNMmwNCxBaqqn9Ed0v9Tfv0CBfDVNm1Va7eG7gjI36W7CPQBwAuBkV8hrKoftOX+ZboLJh8BPHV2fpL/leT+6b4lcw3dm/jNIxb1TboXu7cl2aGte9827xjgVekuWN2R7nD8x9dxVGRiquqXdNce/E2SndJdlPmnrGNbrIej6N4gnkd3rci6nA48KsmydBfMvmF2RpK7JHl6C3m/AK7j19vzfcAbktyvtd05yTPbvM8D90vyuy3YvZyhMDMJSXYDHkMXLh/Ubg+ke0N/QdumnwbekmT7JPelCym9SrJnkke3N8mf0x09md2OVwDLM/c3VLalu/ZiLXBTu5By8CunVwB3avtv1jHAoUlmkuwC/AUL7EtJtm7Ph/fQnbL4yzbrKOCpSZ7Q2myX7iLV3efpO4M+D9w/yYGtr/xv1r+vLGS7DPoE8OR0F/VuA7y61Xryeq5fmyBDx5brK3QXsQ3+ENNJbdrgV2WfQ3ce/1LgeODNVfVfcyz3D+iuWbgaeDO3fEO9K3AcXeA4u9Xw0eEFtDejp9KddriI7nDus9rsI9p9VgEX0r1JvGyexzopL6M78nAB3Xb7WKtng7VAdxrdp72T5mh3AvBx4H/ormf43MDsreiCz6V023s/ugv5qKrj6d7cj033zaKzgCe2eVcBzwTeRnfNyr1oR7wm7PnA6VX1xaq6fPYG/DPwgHRf7fwTuiMhlwNHMs83NTaS29Bti6taHXfm1+Huk+3vD5OcNurO7fTjy+neJH9E9xz47MD8c+hCxgXtNMdSuouWV9Pt1zPp+sJbN7DuvZNcR/d8OpHuYtuHVdWZbb1r6C4ofiPdG/8a4LV0/WadfWfosc32lb+n6yv3bXX/Yr7iFrhdBu9/Ll0ofw/dvnkq8NSqumHeLaNN1uyFbJJ6luQIulNZhy52LdL6aEd8LgaeW1VfXux6tPnxx5ykRZBkOd3XEB+8yKVIc0ryBOAbdKedXkt3TcbXF7UobbY8vSL1LMlf053ueEe7YFfalO1N9xXf2VMcB7brwqQN5ukVSZLUC490SJKkXhg6JElSL3q9kHSXXXap5cuX97lKSZLUo1NPPfWqqrrVP7CEnkPH8uXLWb16dZ+rlCRJPUoy6t8ZAJ5ekSRJPTF0SJKkXhg6JElSLwwdkiSpF4YOSZLUC0OHJEnqhaFDkiT1wtAhSZJ6scn/a/uHvvYji12Chpz6jhcsdgmSpM2QRzokSVIvDB2SJKkXhg5JktQLQ4ckSeqFoUOSJPXC0CFJknph6JAkSb2YN3QkOSLJlUnOGjHv1UkqyS4bpzxJkjQt1udIx5HAAcMTk+wBPB64aMI1SZKkKTRv6KiqVcDVI2a9C3gdUJMuSpIkTZ8FXdOR5OnAJVV1xoTrkSRJU2qD//dKku2BN9KdWlmf9ocAhwAsW7ZsQ1cnSZKmxEKOdNwTuDtwRpLvA7sDpyW566jGVXV4Va2oqhUzMzMLr1SSJG3WNvhIR1WdCdx5drwFjxVVddUE65IkSVNmfb4yewxwCrBnkouTvHDjlyVJkqbNvEc6quo588xfPrFqJEnS1PIXSSVJUi8MHZIkqReGDkmS1AtDhyRJ6oWhQ5Ik9cLQIUmSemHokCRJvTB0SJKkXhg6JElSLwwdkiSpF4YOSZLUC0OHJEnqhaFDkiT1wtAhSZJ6YeiQJEm9MHRIkqRezBs6khyR5MokZw1Me0eSc5L8T5Ljk9x+o1YpSZI2e+tzpONI4IChaScAe1XVA4DvAm+YcF2SJGnKzBs6qmoVcPXQtC9W1U1t9OvA7huhNkmSNEUmcU3HwcB/TGA5kiRpii0Z585J3gTcBBw9R5tDgEMAli1bNs7qtAW56K/uv9glaIRlf3HmYpcgaTO24CMdSVYCTwGeW1W1rnZVdXhVraiqFTMzMwtdnSRJ2swt6EhHkgOA1wH7VdX1ky1JkiRNo/X5yuwxwCnAnkkuTvJC4DBgJ+CEJKcned9GrlOSJG3m5j3SUVXPGTH5gxuhFkmSNMX8RVJJktQLQ4ckSeqFoUOSJPXC0CFJknph6JAkSb0wdEiSpF4YOiRJUi8MHZIkqReGDkmS1AtDhyRJ6oWhQ5Ik9cLQIUmSemHokCRJvTB0SJKkXhg6JElSLwwdkiSpF/OGjiRHJLkyyVkD0+6Y5IQk57W/d9i4ZUqSpM3d+hzpOBI4YGjanwFfqqp7AV9q45IkSes0b+ioqlXA1UOTnw58uA1/GDhwsmVJkqRps9BrOu5SVZe14cuBu0yoHkmSNKWWjLuAqqokta75SQ4BDgFYtmzZuKuTNOX2fc++i12CRvjay7622CVoCiz0SMcVSXYFaH+vXFfDqjq8qlZU1YqZmZkFrk6SJG3uFho6Pgsc1IYPAv51MuVIkqRptT5fmT0GOAXYM8nFSV4IvA14XJLzgMe2cUmSpHWa95qOqnrOOmY9ZsK1SJKkKeYvkkqSpF4YOiRJUi8MHZIkqReGDkmS1AtDhyRJ6oWhQ5Ik9cLQIUmSemHokCRJvTB0SJKkXhg6JElSLwwdkiSpF4YOSZLUC0OHJEnqhaFDkiT1wtAhSZJ6YeiQJEm9GCt0JHlVkm8nOSvJMUm2m1RhkiRpuiw4dCTZDXg5sKKq9gK2Bp49qcIkSdJ0Gff0yhLgtkmWANsDl45fkiRJmkYLDh1VdQnwTuAi4DLgJ1X1xUkVJkmSpss4p1fuADwduDuwFNghyfNGtDskyeokq9euXbvwSiVJ0mZtnNMrjwUurKq1VXUj8Glgn+FGVXV4Va2oqhUzMzNjrE6SJG3OxgkdFwG/nWT7JAEeA5w9mbIkSdK0Geeajm8AxwGnAWe2ZR0+obokSdKUWTLOnavqzcCbJ1SLJEmaYv4iqSRJ6oWhQ5Ik9cLQIUmSemHokCRJvTB0SJKkXhg6JElSLwwdkiSpF4YOSZLUi7F+HEySpEn5yqP2W+wSNMJ+q74ysWV5pEOSJPXC0CFJknph6JAkSb0wdEiSpF4YOiRJUi8MHZIkqReGDkmS1AtDhyRJ6sVYoSPJ7ZMcl+ScJGcn2XtShUmSpOky7i+S/hPwn1X1jCTbAttPoCZJkjSFFhw6kuwMPApYCVBVNwA3TKYsSZI0bcY5vXJ3YC3woST/neQDSXYYbpTkkCSrk6xeu3btGKuTJEmbs3FCxxLgIcD/qaoHAz8F/my4UVUdXlUrqmrFzMzMGKuTJEmbs3FCx8XAxVX1jTZ+HF0IkSRJupUFh46quhxYk2TPNukxwHcmUpUkSZo643575WXA0e2bKxcAfzh+SZIkaRqNFTqq6nRgxWRKkSRJ08xfJJUkSb0wdEiSpF4YOiRJUi8MHZIkqReGDkmS1AtDhyRJ6oWhQ5Ik9cLQIUmSemHokCRJvTB0SJKkXhg6JElSLwwdkiSpF4YOSZLUC0OHJEnqhaFDkiT1wtAhSZJ6MXboSLJ1kv9O8rlJFCRJkqbTJI50vAI4ewLLkSRJU2ys0JFkd+DJwAcmU44kSZpW4x7peDfwOuDm8UuRJEnTbMGhI8lTgCur6tR52h2SZHWS1WvXrl3o6iRJ0mZunCMd+wJPS/J94Fjg0UmOGm5UVYdX1YqqWjEzMzPG6iRJ0uZswaGjqt5QVbtX1XLg2cD/q6rnTawySZI0VfydDkmS1Islk1hIVZ0InDiJZUmSpOnkkQ5JktQLQ4ckSeqFoUOSJPXC0CFJknph6JAkSb0wdEiSpF4YOiRJUi8MHZIkqReGDkmS1AtDhyRJ6oWhQ5Ik9cLQIUmSemHokCRJvTB0SJKkXhg6JElSLwwdkiSpFwsOHUn2SPLlJN9J8u0kr5hkYZIkabosGeO+NwGvrqrTkuwEnJrkhKr6zoRqkyRJU2TBRzqq6rKqOq0NXwucDew2qcIkSdJ0mcg1HUmWAw8GvjGJ5UmSpOkzduhIsiPwKeCVVXXNiPmHJFmdZPXatWvHXZ0kSdpMjRU6kmxDFziOrqpPj2pTVYdX1YqqWjEzMzPO6iRJ0mZsnG+vBPggcHZV/ePkSpIkSdNonCMd+wLPBx6d5PR2e9KE6pIkSVNmwV+ZraqvAplgLZIkaYr5i6SSJKkXhg5JktQLQ4ckSeqFoUOSJPXC0CFJknph6JAkSb0wdEiSpF4YOiRJUi8MHZIkqReGDkmS1AtDhyRJ6oWhQ5Ik9cLQIUmSemHokCRJvTB0SJKkXhg6JElSL8YKHUkOSHJukvOT/NmkipIkSdNnwaEjydbAvwBPBO4LPCfJfSdVmCRJmi7jHOl4OHB+VV1QVTcAxwJPn0xZkiRp2owTOnYD1gyMX9ymSZIk3cqSjb2CJIcAh7TR65Kcu7HXuYnaBbhqsYuYhLzzoMUuYXMzNfueN2exK9jcTM2+z8vd9xtoavY92eB9f7d1zRgndFwC7DEwvnubdgtVdThw+BjrmQpJVlfVisWuQ/1z32+53PdbLvf9aOOcXvkWcK8kd0+yLfBs4LOTKUuSJE2bBR/pqKqbkvwJ8AVga+CIqvr2xCqTJElTZaxrOqrq34F/n1At026LP8W0BXPfb7nc91su9/0IqarFrkGSJG0B/Bl0SZLUC0PHhCQ5uf3dP8nnFrsejS/JW5K8Zo75B67Pr/AmOTLJMyZbnTZ1SVYmWbrYdWjT0PrDYYtdx2IzdExIVe2z2DWodwfS/QsAaZSVgKFDGmDomJAk1w2M3i7J59s/w3tfkq1m2yR5R5JvJ/mvJA9PcmKSC5I8bZFK14Akb0ry3SRfBfZs016c5FtJzkjyqSTbJ9kHeBrwjiSnJ7nnqHYDi35sktVt2U9py12e5KQkp7XbPm36rklWteWeleSRbfrjk5zS2n4yyY79bp0tS5Id2vP4jLYfnpXkMwPzH5fk+CRbt6NZZyU5M8mr2pGtFcDRbT/eNslDk3wlyalJvpBk17acE5O8q/WPs5M8LMmnk5yX5K2L9PC1HtbRRw5Ick57nv6zR76HVJW3CdyA69rf/YGfA/eg+yrxCcAz2rwCntiGjwe+CGwDPBA4fbEfw5Z+Ax4KnAlsD9wOOB94DXCngTZvBV7Who+c3bdtfK52/0kX8u9F9y8Dtmvr2a61uRewug2/GnhTG94a2Inu1w1XATu06a8H/mKxt9k034DfA94/ML4zcA4w08Y/Bjy19ZsTBtrdvv09EVjRhrcBTh6477PofmZgtt3b2/ArgEuBXYHbtL5yp431GL1tlD6ypj2fA3wC+FybtxI4bLFrXuzbRv8Z9C3UN6vqAoAkxwCPAI4DbqB784Huze0XVXVjkjOB5YtRqG7hkcDxVXU9QJLZH7vbq33ivD2wI91v04wyV7tPVNXNwHlJLgDuA1wIHJbkQcAvgXu3tt8CjkiyDfCZqjo9yX50p3K+lu4nibcFThn7EWsuZwL/kOTtdG8cJyX5KPC8JB8C9gZeQBcK75HkPcDn6T5MDNsT2As4oe2/rYHLBubP9rUzgW9X1WUAra/sAfxw0g9OE3GLPgJcC1xYVecBJDmKX/8bENHD/17ZQg1/D3l2/MZqkRe4GfgFQFXdnMR9sek6Ejiwqs5IspLuaNaGthvVJ14FXEF3pGsruiNkVNWqJI8CngwcmeQfgR/RfZp+zvgPR+ujqr6b5CHAk4C3JvkS8AHg3+j21Ser6ibgR0keCDwBeCnw+8DBQ4sLXZjYex2r+0X7e/PA8Oy4rw2bqOE+AnxpkUva5HlNx8bx8HQ/D78V3WHUry52QVovq4AD2/n3negOnUP3SfayduThuQPtr23zmKcdwDOTbJXknnSn3s6lOxR7WTsC8ny6T78kuRtwRVW9n+5N7iHA14F9k/xGa7NDknujjSbdN0+ur6qjgHcAD6mqS+lOfxwKfKi12wXYqqo+1aY/pC1isH+cC8wk2bvdZ5sk9+vtwWijGNFH9gGWt+c5gB8ShpigN45vAYcBvwF8me76DW3iquq0JB8HzgCupNuPAH8OfANY2/7OvpEcC7w/ycuBZ8zRDuAi4Jt014q8tKp+nuS9wKeSvIDutNtPW9v9gdcmuRG4DnhBVa1tR0+OSXKb1u5Q4LuT2wIacn+6C4VvBm4E/qhNP5ru2oyz2/huwIfahwyAN7S/RwLvS/IzulMxzwD+OcnOdK+97wb81xGbt1F9ZBfg80muB07ilq8DWzx/kVSSNkC631r476r64GLXok1bkv2B11TVUxa5lE2GRzokaT0lOZXuiNSrF7sWaXPkkQ5JktQLLySVJEm9MHRIkqReGDokSVIvDB2SJKkXhg5JktQLQ4ckSerF/wdnijLzW8X4lgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for des in ['Scientist', 'Associate', 'Engineer', 'Consultant','Intern', 'Analyst', 'Manager', 'Research', 'Developer','Administrator']:\n",
    "    sub_plot_count = 1\n",
    "    dict_data = {}\n",
    "    for i in df[df['Designation']==des]['most_common_words']:\n",
    "        if i[0][0] in dict_data.keys():\n",
    "            dict_data[i[0][0]] += i[0][1]\n",
    "        else:\n",
    "            dict_data[i[0][0]] = i[0][1]\n",
    "    sorted_x = sorted(dict_data.items(), key=lambda x: x[1], reverse=True)\n",
    "    sorted_dict = collections.OrderedDict(sorted_x)\n",
    "    key = []\n",
    "    value = []\n",
    "    n= 1\n",
    "    for i, j in sorted_dict.items():\n",
    "        key.append(i)\n",
    "        value.append(j)\n",
    "        n += 1\n",
    "        if n==5:\n",
    "            break\n",
    "    plt.figure(figsize=(20, 8))\n",
    "    plt.subplot(2, 2, sub_plot_count+1)\n",
    "    title = \"Words commonly used in \" + des + \" Designation\"\n",
    "    plt.title(title)\n",
    "    sns.barplot(key, value)\n",
    "    sub_plot_count +=1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6bc50be",
   "metadata": {},
   "source": [
    "### Filling missing values of ratings by mean of each Designation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "20475332",
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in ['average_rating_list', 'company_culture','growth_opportunities', 'people_you_work_with', 'person_you_work_for','rewards_you_receive', 'suppport_you_get', 'way_you_work','work_setting']:\n",
    "    df[i] = df[i].fillna(df.groupby('Designation')[i].transform('mean'))\n",
    "    df[i] = df[i].fillna(np.mean(df[i]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "c5787e68",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['final_rating'] = np.round(df.mean(numeric_only=True, axis=1), 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "18c3929e",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['temp_rating'] = np.where(df['final_rating']>=4.0, \"Good\", \"Average\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "13bb9398",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['reviews'].fillna(df['temp_rating'], inplace=True)\n",
    "del df['temp_rating']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "661c2d6f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>reviews</th>\n",
       "      <th>final_rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ensure promotions are given to people who deserve it, change the loyalty gathering process, and make sure all higher ups work in teh field so they can know what works in reality and what does not.</td>\n",
       "      <td>3.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Really great company who puts their employees first. Everyone is so friendly and helpful to their coworkers. The company is great at what it does, helping libraries to access the content they'd like to make available to their patrons. My colleagues were very talented and helped me to build a skillset in the IT field.</td>\n",
       "      <td>4.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Don't be so afraid of coming up with new things and seeing what the development staff can do! If we hadn't broken a few of their IT mandates in 1995, there would have been no support.dell.com worth talking about!!!! Also, don't have all of the developers in IT and crushed by budgets and IT timelines - get them out into the business units so they can be more effective and react faster to changing markets!</td>\n",
       "      <td>3.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>I have worked for Infosys for 10 years in various roles and at various locations. The company has a wealth of resources which can be its strength when you are starting in the company. The large size of the company is a drawback for individual growth once an employee reaches middle-management in the company. I like to work in an environment where processes aid the decision. Having a large company with multiple hierarchy levels though drowns the work into processes and slow down productivity.</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Good</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Average</td>\n",
       "      <td>3.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Stay away from information security. The top level management of this team is causing talent to drop like flies. There is a VP that brought fear-based management style to the organization that has great influencing power at the very top. The VP is the de-facto CISO that runs information security - the actual CISO follows his lead! This has caused strife and there is no unity between teams under the infosec umbrella. Many have been unjustifiably terminated. Many have already have jumped ship in droves. There are a lot of infosec positions currently open at TU. This is not to build up the team, but to back fill open positions. HR is either blind or not paying attention to what is going on because nothing seems to be getting done. Get rid of the bad apples at the top, they are ruining the organization! You've been warned!</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Average</td>\n",
       "      <td>3.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Wake up, see what is in front of your face! You have some good people. The rest are good to smoke cigs and talk, then they say they cannot make money. Try working instead of talking!</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Trust the abilities of the experts you hired to do their job, instead of dictating orders from Korea. Respect our personal time and give us the resources to work smarter, not harder. Money doesn't buy loyalty, reward your employees by recognizing their value to the company. Stop playing favorites!!!</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          reviews  \\\n",
       "0  Ensure promotions are given to people who deserve it, change the loyalty gathering process, and make sure all higher ups work in teh field so they can know what works in reality and what does not.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             \n",
       "1  Really great company who puts their employees first. Everyone is so friendly and helpful to their coworkers. The company is great at what it does, helping libraries to access the content they'd like to make available to their patrons. My colleagues were very talented and helped me to build a skillset in the IT field.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   \n",
       "2  Don't be so afraid of coming up with new things and seeing what the development staff can do! If we hadn't broken a few of their IT mandates in 1995, there would have been no support.dell.com worth talking about!!!! Also, don't have all of the developers in IT and crushed by budgets and IT timelines - get them out into the business units so they can be more effective and react faster to changing markets!                                                                                                                                                                                                                                                                                                                                                                                                                                          \n",
       "3  I have worked for Infosys for 10 years in various roles and at various locations. The company has a wealth of resources which can be its strength when you are starting in the company. The large size of the company is a drawback for individual growth once an employee reaches middle-management in the company. I like to work in an environment where processes aid the decision. Having a large company with multiple hierarchy levels though drowns the work into processes and slow down productivity.                                                                                                                                                                                                                                                                                                                                                  \n",
       "4  Good                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             \n",
       "5  Average                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          \n",
       "6  Stay away from information security. The top level management of this team is causing talent to drop like flies. There is a VP that brought fear-based management style to the organization that has great influencing power at the very top. The VP is the de-facto CISO that runs information security - the actual CISO follows his lead! This has caused strife and there is no unity between teams under the infosec umbrella. Many have been unjustifiably terminated. Many have already have jumped ship in droves. There are a lot of infosec positions currently open at TU. This is not to build up the team, but to back fill open positions. HR is either blind or not paying attention to what is going on because nothing seems to be getting done. Get rid of the bad apples at the top, they are ruining the organization! You've been warned!   \n",
       "7  Average                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          \n",
       "8  Wake up, see what is in front of your face! You have some good people. The rest are good to smoke cigs and talk, then they say they cannot make money. Try working instead of talking!                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           \n",
       "9   Trust the abilities of the experts you hired to do their job, instead of dictating orders from Korea. Respect our personal time and give us the resources to work smarter, not harder. Money doesn't buy loyalty, reward your employees by recognizing their value to the company. Stop playing favorites!!!                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    \n",
       "\n",
       "   final_rating  \n",
       "0  3.6           \n",
       "1  4.8           \n",
       "2  3.9           \n",
       "3  4.0           \n",
       "4  4.0           \n",
       "5  3.9           \n",
       "6  4.0           \n",
       "7  3.9           \n",
       "8  3.0           \n",
       "9  4.0           "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['reviews', 'final_rating']].head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "66e510b6",
   "metadata": {},
   "source": [
    "### Performing Sentiment Analysis on reviews column "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "70e047f3",
   "metadata": {},
   "outputs": [],
   "source": [
    "sid_obj = SentimentIntensityAnalyzer()\n",
    "polarity = []\n",
    "for i in df[\"reviews\"]:\n",
    "    polarity.append(sid_obj.polarity_scores(i)['compound'])\n",
    "polarity = np.array(polarity)\n",
    "df['polarity']=polarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "7cb467ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "col         = 'polarity'\n",
    "conditions  = [df[col] > 0.5, df[col] == 0.5, df[col] < 0.5]\n",
    "choices     = [ \"Positive\", 'Neutral', 'Negative' ]\n",
    "    \n",
    "df[\"Sentiment\"] = np.select(conditions, choices, default=np.nan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "725c0eb4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>reviews</th>\n",
       "      <th>Designation</th>\n",
       "      <th>final_rating</th>\n",
       "      <th>polarity</th>\n",
       "      <th>Sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ensure promotions are given to people who deserve it, change the loyalty gathering process, and make sure all higher ups work in teh field so they can know what works in reality and what does not.</td>\n",
       "      <td>Scientist</td>\n",
       "      <td>3.6</td>\n",
       "      <td>0.8126</td>\n",
       "      <td>Positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Really great company who puts their employees first. Everyone is so friendly and helpful to their coworkers. The company is great at what it does, helping libraries to access the content they'd like to make available to their patrons. My colleagues were very talented and helped me to build a skillset in the IT field.</td>\n",
       "      <td>Associate</td>\n",
       "      <td>4.8</td>\n",
       "      <td>0.9757</td>\n",
       "      <td>Positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Don't be so afraid of coming up with new things and seeing what the development staff can do! If we hadn't broken a few of their IT mandates in 1995, there would have been no support.dell.com worth talking about!!!! Also, don't have all of the developers in IT and crushed by budgets and IT timelines - get them out into the business units so they can be more effective and react faster to changing markets!</td>\n",
       "      <td>Engineer</td>\n",
       "      <td>3.9</td>\n",
       "      <td>0.6143</td>\n",
       "      <td>Positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>I have worked for Infosys for 10 years in various roles and at various locations. The company has a wealth of resources which can be its strength when you are starting in the company. The large size of the company is a drawback for individual growth once an employee reaches middle-management in the company. I like to work in an environment where processes aid the decision. Having a large company with multiple hierarchy levels though drowns the work into processes and slow down productivity.</td>\n",
       "      <td>Consultant</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.8176</td>\n",
       "      <td>Positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Good</td>\n",
       "      <td>Other</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.4404</td>\n",
       "      <td>Negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>385</th>\n",
       "      <td>You should find a way to compensate employees who give the extra mile to avoid the losing of enthusiasm of the employee to move forward with the company. Also the company should give (minimum) a course of human relationship tuition to the supervisors and managers.</td>\n",
       "      <td>Analyst</td>\n",
       "      <td>3.7</td>\n",
       "      <td>-0.2263</td>\n",
       "      <td>Negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>386</th>\n",
       "      <td>Good</td>\n",
       "      <td>Associate</td>\n",
       "      <td>4.1</td>\n",
       "      <td>0.4404</td>\n",
       "      <td>Negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>387</th>\n",
       "      <td>Good</td>\n",
       "      <td>Engineer</td>\n",
       "      <td>4.1</td>\n",
       "      <td>0.4404</td>\n",
       "      <td>Negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>388</th>\n",
       "      <td>Good</td>\n",
       "      <td>Engineer</td>\n",
       "      <td>4.1</td>\n",
       "      <td>0.4404</td>\n",
       "      <td>Negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>389</th>\n",
       "      <td>This is an outreach program organized by Google. It's main purpose is to find talented college/university students who are passionate about technology and are knowledgeable about it. These students will be picked to become Google representatives (not employees) in their respective colleges/universities with the following responsibilities:\\nPromote, teach and help adopt Google Services to students, teachers and businesses.\\nBecome point of contact between Google and those mentioned above.\\nOrganize events inside your respective colleges/universities to either promote the services, teach how to use them or at least use Google's services to help minimize costs and make the event a success.\\nYou are there to learn and show the student body that Google is more than a technology company.</td>\n",
       "      <td>Engineer</td>\n",
       "      <td>4.2</td>\n",
       "      <td>0.9806</td>\n",
       "      <td>Positive</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>390 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      reviews  \\\n",
       "0    Ensure promotions are given to people who deserve it, change the loyalty gathering process, and make sure all higher ups work in teh field so they can know what works in reality and what does not.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       \n",
       "1    Really great company who puts their employees first. Everyone is so friendly and helpful to their coworkers. The company is great at what it does, helping libraries to access the content they'd like to make available to their patrons. My colleagues were very talented and helped me to build a skillset in the IT field.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             \n",
       "2    Don't be so afraid of coming up with new things and seeing what the development staff can do! If we hadn't broken a few of their IT mandates in 1995, there would have been no support.dell.com worth talking about!!!! Also, don't have all of the developers in IT and crushed by budgets and IT timelines - get them out into the business units so they can be more effective and react faster to changing markets!                                                                                                                                                                                                                                                                                                                                                                                                    \n",
       "3    I have worked for Infosys for 10 years in various roles and at various locations. The company has a wealth of resources which can be its strength when you are starting in the company. The large size of the company is a drawback for individual growth once an employee reaches middle-management in the company. I like to work in an environment where processes aid the decision. Having a large company with multiple hierarchy levels though drowns the work into processes and slow down productivity.                                                                                                                                                                                                                                                                                                            \n",
       "4    Good                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       \n",
       "..    ...                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       \n",
       "385  You should find a way to compensate employees who give the extra mile to avoid the losing of enthusiasm of the employee to move forward with the company. Also the company should give (minimum) a course of human relationship tuition to the supervisors and managers.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   \n",
       "386  Good                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       \n",
       "387  Good                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       \n",
       "388  Good                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       \n",
       "389  This is an outreach program organized by Google. It's main purpose is to find talented college/university students who are passionate about technology and are knowledgeable about it. These students will be picked to become Google representatives (not employees) in their respective colleges/universities with the following responsibilities:\\nPromote, teach and help adopt Google Services to students, teachers and businesses.\\nBecome point of contact between Google and those mentioned above.\\nOrganize events inside your respective colleges/universities to either promote the services, teach how to use them or at least use Google's services to help minimize costs and make the event a success.\\nYou are there to learn and show the student body that Google is more than a technology company.   \n",
       "\n",
       "    Designation  final_rating  polarity Sentiment  \n",
       "0    Scientist   3.6           0.8126    Positive  \n",
       "1    Associate   4.8           0.9757    Positive  \n",
       "2    Engineer    3.9           0.6143    Positive  \n",
       "3    Consultant  4.0           0.8176    Positive  \n",
       "4    Other       4.0           0.4404    Negative  \n",
       "..     ...       ...              ...         ...  \n",
       "385  Analyst     3.7          -0.2263    Negative  \n",
       "386  Associate   4.1           0.4404    Negative  \n",
       "387  Engineer    4.1           0.4404    Negative  \n",
       "388  Engineer    4.1           0.4404    Negative  \n",
       "389  Engineer    4.2           0.9806    Positive  \n",
       "\n",
       "[390 rows x 5 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[:, -5:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "ea9ae486",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Job_title', 'company_name', 'company_location', 'City', 'State',\n",
       "       'job_type', 'posted_date', 'qualification', 'benefits', 'description',\n",
       "       'desc_tokenized', 'no_stopwords', 'desc_stemmed', 'desc_lemmatized',\n",
       "       'most_common_words', 'average_rating_list', 'company_culture',\n",
       "       'growth_opportunities', 'people_you_work_with', 'person_you_work_for',\n",
       "       'rewards_you_receive', 'suppport_you_get', 'way_you_work',\n",
       "       'work_setting', 'reviews', 'Designation', 'final_rating', 'polarity',\n",
       "       'Sentiment'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "01ce8f27",
   "metadata": {},
   "outputs": [],
   "source": [
    "degree_list = []\n",
    "df['qualification'].str.contains('')\n",
    "for i in df['qualification']:\n",
    "    degree_sub_list = []\n",
    "    for j in i[1:-1].split(','):\n",
    "        if 'degree' in j.lower():\n",
    "            degree_sub_list.append(j)\n",
    "    else:\n",
    "        degree_list.append(\"Not mentioned\")\n",
    "    degree_list.append(degree_sub_list)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "362bd059",
   "metadata": {},
   "outputs": [],
   "source": [
    "#np.where(df['qualification'].str.contains('degree'), \"Yes\", \"No\")\n",
    "df[df['qualification'].str.contains('degree')]\n",
    "\n",
    "word_list = [\"Bachelor's degree\", \"Master's degree\", \"Doctoral degree\"]\n",
    "df['degree'] = df.qualification.str.extract('({})'.format('|'.join(word_list)), flags=re.IGNORECASE, expand=False).str.lower().fillna('Not mentioned')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "95d81404",
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in ['average_rating_list', 'company_culture', 'growth_opportunities', 'people_you_work_with', 'person_you_work_for', 'rewards_you_receive', 'suppport_you_get', 'way_you_work', 'work_setting', 'final_rating']:\n",
    "    df[i] = np.round(df[i], 1)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "1ea03af1",
   "metadata": {},
   "outputs": [],
   "source": [
    "df[['Job_title', 'company_name', 'City', 'State', 'job_type', 'qualification', 'average_rating_list', 'company_culture', 'growth_opportunities', 'people_you_work_with', 'person_you_work_for', 'rewards_you_receive', 'suppport_you_get', 'way_you_work', 'work_setting', 'Designation', 'final_rating', 'polarity',\n",
    "       'Sentiment']].to_csv('dashbaord_data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "722e718d",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}