Word2vec.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Word2vec model\n",
    "\n",
    "Word2Vec is one of the most popular and more advanced techniques for learning word embeddings. This technique is based on the assumption that words in the same contexts usually have similar meanings. Word embedding is a vector representation of a word and its input is the text corpus, and its output is a set of vectors. Word embeddings using word2vec can make natural language readable for the computer, and then one can apply further implementation of mathematical operations on words to detect their similarity. A well-trained set of word vectors will place similar words close together in this space.\n",
    "\n",
    "In this project, I would like to show how to use the Gensim implementation of Word2Vec by using python with an example of generating word embedding for the make-up products model.\n",
    "This dataset include such features as name of product, description and tags. I used these functions to generate word embedding for each those example and then compare similarities between them. Before I started generate Word2vec model I had to pre-process data. I applied process in which sentences are converted to a list of words, removed stopwords and  applied lemmatizing. Then I created Word2Vec model for each example. Finally I used T-SNE to visualize received word embeddings. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Importing packages and loading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "import seaborn as sns\n",
    "sns.set_style(\"darkgrid\")\n",
    "\n",
    "import re\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.stem.wordnet import WordNetLemmatizer\n",
    "\n",
    "import gensim\n",
    "from gensim.utils import simple_preprocess\n",
    "from gensim.models import phrases, word2vec, Word2Vec\n",
    "from gensim.models.phrases import Phrases, Phraser\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.manifold import TSNE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "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>brand</th>\n",
       "      <th>category</th>\n",
       "      <th>currency</th>\n",
       "      <th>description</th>\n",
       "      <th>id</th>\n",
       "      <th>name</th>\n",
       "      <th>price</th>\n",
       "      <th>price_sign</th>\n",
       "      <th>product_type</th>\n",
       "      <th>rating</th>\n",
       "      <th>tag_list</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>colourpop</td>\n",
       "      <td>pencil</td>\n",
       "      <td>CAD</td>\n",
       "      <td>Lippie Pencil A long-wearing and high-intensit...</td>\n",
       "      <td>1048</td>\n",
       "      <td>Lippie Pencil</td>\n",
       "      <td>5.0</td>\n",
       "      <td>$</td>\n",
       "      <td>lip_liner</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Vegan', 'cruelty free']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>colourpop</td>\n",
       "      <td>lipstick</td>\n",
       "      <td>CAD</td>\n",
       "      <td>Blotted Lip Sheer matte lipstick that creates ...</td>\n",
       "      <td>1047</td>\n",
       "      <td>Blotted Lip</td>\n",
       "      <td>5.5</td>\n",
       "      <td>$</td>\n",
       "      <td>lipstick</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Vegan', 'cruelty free']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>colourpop</td>\n",
       "      <td>lipstick</td>\n",
       "      <td>CAD</td>\n",
       "      <td>Lippie Stix Formula contains Vitamin E, Mango,...</td>\n",
       "      <td>1046</td>\n",
       "      <td>Lippie Stix</td>\n",
       "      <td>5.5</td>\n",
       "      <td>$</td>\n",
       "      <td>lipstick</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Vegan', 'cruelty free']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>colourpop</td>\n",
       "      <td>liquid</td>\n",
       "      <td>CAD</td>\n",
       "      <td>Developed for the Selfie Age, our buildable fu...</td>\n",
       "      <td>1045</td>\n",
       "      <td>No Filter Foundation</td>\n",
       "      <td>12.0</td>\n",
       "      <td>$</td>\n",
       "      <td>foundation</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Vegan', 'cruelty free']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>boosh</td>\n",
       "      <td>lipstick</td>\n",
       "      <td>CAD</td>\n",
       "      <td>All of our products are free from lead and hea...</td>\n",
       "      <td>1044</td>\n",
       "      <td>Lipstick</td>\n",
       "      <td>26.0</td>\n",
       "      <td>$</td>\n",
       "      <td>lipstick</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Organic', 'Chemical Free']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       brand  category currency  \\\n",
       "0  colourpop    pencil      CAD   \n",
       "1  colourpop  lipstick      CAD   \n",
       "2  colourpop  lipstick      CAD   \n",
       "3  colourpop    liquid      CAD   \n",
       "4      boosh  lipstick      CAD   \n",
       "\n",
       "                                         description    id  \\\n",
       "0  Lippie Pencil A long-wearing and high-intensit...  1048   \n",
       "1  Blotted Lip Sheer matte lipstick that creates ...  1047   \n",
       "2  Lippie Stix Formula contains Vitamin E, Mango,...  1046   \n",
       "3  Developed for the Selfie Age, our buildable fu...  1045   \n",
       "4  All of our products are free from lead and hea...  1044   \n",
       "\n",
       "                   name  price price_sign product_type  rating  \\\n",
       "0         Lippie Pencil    5.0          $    lip_liner     NaN   \n",
       "1           Blotted Lip    5.5          $     lipstick     NaN   \n",
       "2           Lippie Stix    5.5          $     lipstick     NaN   \n",
       "3  No Filter Foundation   12.0          $   foundation     NaN   \n",
       "4              Lipstick   26.0          $     lipstick     NaN   \n",
       "\n",
       "                       tag_list  \n",
       "0     ['Vegan', 'cruelty free']  \n",
       "1     ['Vegan', 'cruelty free']  \n",
       "2     ['Vegan', 'cruelty free']  \n",
       "3     ['Vegan', 'cruelty free']  \n",
       "4  ['Organic', 'Chemical Free']  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('C:\\\\Python Scripts\\\\API_products\\\\products_final.csv', header=0,index_col=0)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 931 entries, 0 to 930\n",
      "Data columns (total 11 columns):\n",
      "brand           919 non-null object\n",
      "category        507 non-null object\n",
      "currency        368 non-null object\n",
      "description     906 non-null object\n",
      "id              931 non-null int64\n",
      "name            931 non-null object\n",
      "price           917 non-null float64\n",
      "price_sign      368 non-null object\n",
      "product_type    931 non-null object\n",
      "rating          340 non-null float64\n",
      "tag_list        931 non-null object\n",
      "dtypes: float64(2), int64(1), object(8)\n",
      "memory usage: 87.3+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(931, 11)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Preparation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "text = df.drop(['product_type', 'currency', 'id', 'price', 'price_sign', 'rating', 'brand', 'category'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>description</th>\n",
       "      <th>name</th>\n",
       "      <th>tag_list</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Lippie Pencil A long-wearing and high-intensit...</td>\n",
       "      <td>Lippie Pencil</td>\n",
       "      <td>['Vegan', 'cruelty free']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Blotted Lip Sheer matte lipstick that creates ...</td>\n",
       "      <td>Blotted Lip</td>\n",
       "      <td>['Vegan', 'cruelty free']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Lippie Stix Formula contains Vitamin E, Mango,...</td>\n",
       "      <td>Lippie Stix</td>\n",
       "      <td>['Vegan', 'cruelty free']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Developed for the Selfie Age, our buildable fu...</td>\n",
       "      <td>No Filter Foundation</td>\n",
       "      <td>['Vegan', 'cruelty free']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>All of our products are free from lead and hea...</td>\n",
       "      <td>Lipstick</td>\n",
       "      <td>['Organic', 'Chemical Free']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         description                  name  \\\n",
       "0  Lippie Pencil A long-wearing and high-intensit...         Lippie Pencil   \n",
       "1  Blotted Lip Sheer matte lipstick that creates ...           Blotted Lip   \n",
       "2  Lippie Stix Formula contains Vitamin E, Mango,...           Lippie Stix   \n",
       "3  Developed for the Selfie Age, our buildable fu...  No Filter Foundation   \n",
       "4  All of our products are free from lead and hea...              Lipstick   \n",
       "\n",
       "                       tag_list  \n",
       "0     ['Vegan', 'cruelty free']  \n",
       "1     ['Vegan', 'cruelty free']  \n",
       "2     ['Vegan', 'cruelty free']  \n",
       "3     ['Vegan', 'cruelty free']  \n",
       "4  ['Organic', 'Chemical Free']  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "text['description'] = text['description'].astype(str)\n",
    "text['tag_list'] = text['tag_list'].astype(str)\n",
    "text['name'] = text['name'].astype(str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "description    0\n",
       "name           0\n",
       "tag_list       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Text preprocessing\n",
    "\n",
    "We are removing non-alphabetic characters, the stopwords and lemmatizing for each line of text."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# This function converts a text to a sequence of words:\n",
    "def tokens(words):\n",
    "    words = re.sub(\"[^a-zA-Z]\",\" \",words)\n",
    "    text = words.lower().split()\n",
    "    return \" \".join(text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>description</th>\n",
       "      <th>name</th>\n",
       "      <th>tag_list</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>lippie pencil a long wearing and high intensit...</td>\n",
       "      <td>lippie pencil</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>blotted lip sheer matte lipstick that creates ...</td>\n",
       "      <td>blotted lip</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>lippie stix formula contains vitamin e mango a...</td>\n",
       "      <td>lippie stix</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>developed for the selfie age our buildable ful...</td>\n",
       "      <td>no filter foundation</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>all of our products are free from lead and hea...</td>\n",
       "      <td>lipstick</td>\n",
       "      <td>organic chemical free</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         description                  name  \\\n",
       "0  lippie pencil a long wearing and high intensit...         lippie pencil   \n",
       "1  blotted lip sheer matte lipstick that creates ...           blotted lip   \n",
       "2  lippie stix formula contains vitamin e mango a...           lippie stix   \n",
       "3  developed for the selfie age our buildable ful...  no filter foundation   \n",
       "4  all of our products are free from lead and hea...              lipstick   \n",
       "\n",
       "                tag_list  \n",
       "0     vegan cruelty free  \n",
       "1     vegan cruelty free  \n",
       "2     vegan cruelty free  \n",
       "3     vegan cruelty free  \n",
       "4  organic chemical free  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text['description'] = text['description'].apply(tokens)\n",
    "text['tag_list'] = text['tag_list'].apply(tokens)\n",
    "text['name'] = text['name'].apply(tokens)\n",
    "text.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### StopWords\n",
    "\n",
    "We extracting the stopwords from nltk library, then we build function to remove stop words:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', \"you're\"]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show some stop words\n",
    "stop = stopwords.words('english')\n",
    "stop[0:10]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def stopwords(text):\n",
    "    text = [word.lower() for word in text.split() if word.lower() not in stop]\n",
    "    return \" \".join(text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>description</th>\n",
       "      <th>name</th>\n",
       "      <th>tag_list</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>lippie pencil long wearing high intensity lip ...</td>\n",
       "      <td>lippie pencil</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>blotted lip sheer matte lipstick creates perfe...</td>\n",
       "      <td>blotted lip</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>lippie stix formula contains vitamin e mango a...</td>\n",
       "      <td>lippie stix</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>developed selfie age buildable full coverage n...</td>\n",
       "      <td>filter foundation</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>products free lead heavy metals parabens phtha...</td>\n",
       "      <td>lipstick</td>\n",
       "      <td>organic chemical free</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         description               name  \\\n",
       "0  lippie pencil long wearing high intensity lip ...      lippie pencil   \n",
       "1  blotted lip sheer matte lipstick creates perfe...        blotted lip   \n",
       "2  lippie stix formula contains vitamin e mango a...        lippie stix   \n",
       "3  developed selfie age buildable full coverage n...  filter foundation   \n",
       "4  products free lead heavy metals parabens phtha...           lipstick   \n",
       "\n",
       "                tag_list  \n",
       "0     vegan cruelty free  \n",
       "1     vegan cruelty free  \n",
       "2     vegan cruelty free  \n",
       "3     vegan cruelty free  \n",
       "4  organic chemical free  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text['description'] = text['description'].apply(stopwords)\n",
    "text['name'] = text['name'].apply(stopwords)\n",
    "text.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We build the function to apply lemmatizing:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "lem = WordNetLemmatizer()\n",
    "\n",
    "def word_lem(text):\n",
    "    lem_text = [lem.lemmatize(word) for word in text.split()]\n",
    "    return \" \".join(lem_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>description</th>\n",
       "      <th>name</th>\n",
       "      <th>tag_list</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>lippie pencil long wearing high intensity lip ...</td>\n",
       "      <td>lippie pencil</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>blotted lip sheer matte lipstick creates perfe...</td>\n",
       "      <td>blotted lip</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>lippie stix formula contains vitamin e mango a...</td>\n",
       "      <td>lippie stix</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>developed selfie age buildable full coverage n...</td>\n",
       "      <td>filter foundation</td>\n",
       "      <td>vegan cruelty free</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>product free lead heavy metal parabens phthala...</td>\n",
       "      <td>lipstick</td>\n",
       "      <td>organic chemical free</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         description               name  \\\n",
       "0  lippie pencil long wearing high intensity lip ...      lippie pencil   \n",
       "1  blotted lip sheer matte lipstick creates perfe...        blotted lip   \n",
       "2  lippie stix formula contains vitamin e mango a...        lippie stix   \n",
       "3  developed selfie age buildable full coverage n...  filter foundation   \n",
       "4  product free lead heavy metal parabens phthala...           lipstick   \n",
       "\n",
       "                tag_list  \n",
       "0     vegan cruelty free  \n",
       "1     vegan cruelty free  \n",
       "2     vegan cruelty free  \n",
       "3     vegan cruelty free  \n",
       "4  organic chemical free  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text['description']  = text['description'].apply(word_lem)\n",
    "text.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'blotted lip sheer matte lipstick creates perfect popsicle pout formula lightweight matte buildable light medium coverage'"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text['description'][1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model creation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Word2vec for name\n",
    "\n",
    "Lets try to use simple_preprocess method, which convert a document into a list of lowercase tokens, ignoring tokens that are too short or too long."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "corpus = text['name'].map(simple_preprocess)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                         [lippie, pencil]\n",
       "1                           [blotted, lip]\n",
       "2                           [lippie, stix]\n",
       "3                     [filter, foundation]\n",
       "4                               [lipstick]\n",
       "5                      [serum, foundation]\n",
       "6                   [coverage, foundation]\n",
       "7                          [liquid, liner]\n",
       "8                              [eyeshadow]\n",
       "9    [realist, invisible, setting, powder]\n",
       "Name: name, dtype: object"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corpus[0:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can train the genism word2vec model with our own custom corpus as following:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PC\\Anaconda3\\lib\\site-packages\\gensim\\models\\base_any2vec.py:743: UserWarning: C extension not loaded, training will be slow. Install a C compiler and reinstall gensim for fast training.\n",
      "  \"C extension not loaded, training will be slow. \"\n"
     ]
    }
   ],
   "source": [
    "name_model = Word2Vec(corpus, size=100, window=3, min_count=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('rose', 0.3261679410934448),\n",
       " ('maybelline', 0.30425405502319336),\n",
       " ('express', 0.30156373977661133),\n",
       " ('long', 0.2996756434440613),\n",
       " ('bronzer', 0.2945731580257416),\n",
       " ('nail', 0.29042112827301025),\n",
       " ('flat', 0.29022955894470215),\n",
       " ('flexible', 0.28989794850349426),\n",
       " ('nude', 0.2737104594707489),\n",
       " ('million', 0.27251937985420227)]"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# This will print the most similar words present in the model:\n",
    "name_model.wv.most_similar('lipstick')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('stick', 0.3793315291404724),\n",
       " ('metals', 0.33757591247558594),\n",
       " ('contour', 0.31963130831718445),\n",
       " ('boudoir', 0.3114571273326874),\n",
       " ('addict', 0.3113562762737274),\n",
       " ('rouge', 0.29319578409194946),\n",
       " ('brilliant', 0.2928236722946167),\n",
       " ('gel', 0.28807365894317627),\n",
       " ('resistant', 0.280072420835495),\n",
       " ('fail', 0.27691227197647095)]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name_model.wv.most_similar('liquid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('spectrum', 0.31646865606307983),\n",
       " ('radiance', 0.3039230406284332),\n",
       " ('transforming', 0.30042147636413574),\n",
       " ('muse', 0.29501938819885254),\n",
       " ('matte', 0.2898262143135071),\n",
       " ('paris', 0.28061944246292114),\n",
       " ('kohl', 0.27684545516967773),\n",
       " ('cushion', 0.2622339725494385),\n",
       " ('plumping', 0.25422000885009766),\n",
       " ('exposure', 0.253739595413208)]"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name_model.wv.most_similar('pencil')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Word2vec model for description"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# We create the list of list format of the custom corpus for gensim modeling:\n",
    "sentences = [row.split() for row in text['description']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['lippie',\n",
       "  'pencil',\n",
       "  'long',\n",
       "  'wearing',\n",
       "  'high',\n",
       "  'intensity',\n",
       "  'lip',\n",
       "  'pencil',\n",
       "  'glide',\n",
       "  'easily',\n",
       "  'prevents',\n",
       "  'feathering',\n",
       "  'many',\n",
       "  'lippie',\n",
       "  'stix',\n",
       "  'coordinating',\n",
       "  'lippie',\n",
       "  'pencil',\n",
       "  'designed',\n",
       "  'compliment',\n",
       "  'perfectly',\n",
       "  'feel',\n",
       "  'free',\n",
       "  'mix',\n",
       "  'match']]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sentences[0:1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Creating the model and setting values for the various parameters:\n",
    "\n",
    "Here are few important of the hyperparameters of this model:\n",
    "- size: this is the number of dimensions of the embeddings. Typical numbers is in the range of 50 to 300, we use 100 beacause we don't have that much text.\n",
    "- window: it is the maximum distance between a target word and words around the target word.\n",
    "- min_count:  minimum count of words to consider when training the model. Words with occurrence less than this count will be ignored. We used for min_count 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PC\\Anaconda3\\lib\\site-packages\\gensim\\models\\base_any2vec.py:743: UserWarning: C extension not loaded, training will be slow. Install a C compiler and reinstall gensim for fast training.\n",
      "  \"C extension not loaded, training will be slow. \"\n"
     ]
    }
   ],
   "source": [
    "model = Word2Vec(sentences, min_count=1, size=100, window=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('light', 0.9998731017112732),\n",
       " ('new', 0.9998553991317749),\n",
       " ('bold', 0.9998409152030945),\n",
       " ('vibrant', 0.9997850656509399),\n",
       " ('available', 0.9997632503509521),\n",
       " ('complexion', 0.9997614622116089),\n",
       " ('glossy', 0.9997571110725403),\n",
       " ('blendable', 0.999735951423645),\n",
       " ('provide', 0.9997149705886841),\n",
       " ('nail', 0.9997074604034424)]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#This will print the most similar words present in the model: \n",
    "model.wv.most_similar('lipstick')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('essential', 0.9990046620368958),\n",
       " ('silk', 0.9988724589347839),\n",
       " ('undertone', 0.9988225698471069),\n",
       " ('complex', 0.9987339973449707),\n",
       " ('rose', 0.9986394643783569),\n",
       " ('certified', 0.9986120462417603),\n",
       " ('panthenol', 0.9985828399658203),\n",
       " ('precious', 0.9985687732696533),\n",
       " ('gold', 0.9984335899353027),\n",
       " ('contains', 0.9984298944473267)]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv.most_similar('powder')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('package', 0.976388692855835),\n",
       " ('date', 0.9733823537826538),\n",
       " ('receive', 0.972027063369751),\n",
       " ('ingredient', 0.9605628848075867),\n",
       " ('time', 0.9588063359260559),\n",
       " ('isododecane', 0.9498820900917053),\n",
       " ('twisted', 0.947334349155426),\n",
       " ('talc', 0.9453209042549133),\n",
       " ('list', 0.9449949264526367),\n",
       " ('confidence', 0.9344406127929688)]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv.most_similar('product')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('cream', 0.9998112916946411),\n",
       " ('perfect', 0.9997818470001221),\n",
       " ('easy', 0.9997705221176147),\n",
       " ('lightweight', 0.99977046251297),\n",
       " ('tone', 0.9997384548187256),\n",
       " ('creamy', 0.9997354745864868),\n",
       " ('skin', 0.9997190237045288),\n",
       " ('soft', 0.9996866583824158),\n",
       " ('rich', 0.9996604323387146),\n",
       " ('colour', 0.9996539354324341)]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv.most_similar('blush')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can use Word2vec to calculate the similarity between two words:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9984949"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv.similarity('bronzer', 'blush')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.83895385"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv.similarity('product', 'powder')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PC\\Anaconda3\\lib\\site-packages\\gensim\\models\\keyedvectors.py:877: FutureWarning: arrays to stack must be passed as a \"sequence\" type such as list or tuple. Support for non-sequence iterables such as generators is deprecated as of NumPy 1.16 and will raise an error in the future.\n",
      "  vectors = vstack(self.word_vec(word, use_norm=True) for word in used_words).astype(REAL)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'book'"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#This will print the odd word among them:\n",
    "model.wv.doesnt_match([\"blush\", \"lipstick\", \"book\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PC\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:2: DeprecationWarning: Call to deprecated `syn0` (Attribute will be removed in 4.0.0, use self.vectors instead).\n",
      "  \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(5016, 100)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# This will give the total number of words in the vocabolary created from this dataset:\n",
    "model.wv.syn0.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('word2vec.model')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Bigrams"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We used Gensim Phrases package to automatically detect common phrases (bigrams) from a list of sentences:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "bigrams = phrases.Phrases(sentences)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PC\\Anaconda3\\lib\\site-packages\\gensim\\models\\base_any2vec.py:743: UserWarning: C extension not loaded, training will be slow. Install a C compiler and reinstall gensim for fast training.\n",
      "  \"C extension not loaded, training will be slow. \"\n"
     ]
    }
   ],
   "source": [
    "bigrams[sentences]\n",
    "\n",
    "bigram_model = word2vec.Word2Vec(bigrams[sentences], size=100, min_count=3, iter=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('healthy_looking', 0.9951655864715576),\n",
       " ('illuminating', 0.9928097128868103),\n",
       " ('veil', 0.992239236831665),\n",
       " ('dot', 0.9901595115661621),\n",
       " ('light_medium', 0.990126371383667),\n",
       " ('glowing', 0.9897735714912415),\n",
       " ('four', 0.9892417192459106),\n",
       " ('high_definition', 0.9891389608383179),\n",
       " ('pressed_powder', 0.989067792892456),\n",
       " ('delicate', 0.9881424307823181)]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bigram_model.wv.most_similar('flush_color')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('veil', 0.9951770901679993),\n",
       " ('flush_color', 0.9951655864715576),\n",
       " ('pressed_powder', 0.9948857426643372),\n",
       " ('light_medium', 0.994127631187439),\n",
       " ('illuminating', 0.9919139742851257),\n",
       " ('high_definition', 0.9914615154266357),\n",
       " ('perfection', 0.9913911819458008),\n",
       " ('enough', 0.9907406568527222),\n",
       " ('allowing', 0.9901889562606812),\n",
       " ('fluid', 0.9901399612426758)]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bigram_model.wv.most_similar('healthy_looking')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('tint', 0.9947993159294128),\n",
       " ('leaving', 0.9945440292358398),\n",
       " ('whipped', 0.9940870404243469),\n",
       " ('wash', 0.993788480758667),\n",
       " ('alone', 0.9937129020690918),\n",
       " ('level', 0.992255449295044),\n",
       " ('refreshingly', 0.9914376735687256),\n",
       " ('pressed_powder', 0.9913636445999146),\n",
       " ('combine', 0.9913498759269714),\n",
       " ('perfection', 0.99062579870224)]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bigram_model.wv.most_similar('silky_smooth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('allowing', 0.9951421618461609),\n",
       " ('perfection', 0.9951174259185791),\n",
       " ('healthy_looking', 0.9948857426643372),\n",
       " ('veil', 0.993191659450531),\n",
       " ('level', 0.9925819039344788),\n",
       " ('e_l', 0.9925537109375),\n",
       " ('fluid', 0.9923195838928223),\n",
       " ('fit', 0.9921626448631287),\n",
       " ('alone', 0.9915205240249634),\n",
       " ('silky_smooth', 0.9913636445999146)]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bigram_model.wv.most_similar('pressed_powder')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.99136376"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bigram_model.wv.similarity('pressed_powder', 'silky_smooth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9774893"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bigram_model.wv.similarity('healthy_looking', 'multi_dimensional')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then, given positive and negative words, we will find two best words that are similar to positive and opposite to negative words:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('alone', 0.996069073677063), ('wash', 0.9956036806106567)]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w1 = [\"silky_smooth\",\"healthy_looking\"]\n",
    "w2 = [\"flush_color\"]\n",
    "bigram_model.wv.most_similar (positive=w1,negative=w2,topn=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PC\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:1: DeprecationWarning: Call to deprecated `syn0` (Attribute will be removed in 4.0.0, use self.vectors instead).\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(2778, 100)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bigram_model.wv.syn0.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Save the word2vec model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "bigram_model.save('word2vec_bigram.model')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Word2vec for Tags"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "sent = [row.split() for row in text['tag_list']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['vegan', 'cruelty', 'free'],\n",
       " ['vegan', 'cruelty', 'free'],\n",
       " ['vegan', 'cruelty', 'free'],\n",
       " ['vegan', 'cruelty', 'free'],\n",
       " ['organic', 'chemical', 'free']]"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sent[0:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PC\\Anaconda3\\lib\\site-packages\\gensim\\models\\base_any2vec.py:743: UserWarning: C extension not loaded, training will be slow. Install a C compiler and reinstall gensim for fast training.\n",
      "  \"C extension not loaded, training will be slow. \"\n"
     ]
    }
   ],
   "source": [
    "tag_model = Word2Vec(sent, min_count=1, size=50, window=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('alcohol', 0.23045749962329865),\n",
       " ('cruelty', 0.21317705512046814),\n",
       " ('dairy', 0.19818338751792908),\n",
       " ('usda', 0.18433375656604767),\n",
       " ('gluten', 0.18179476261138916),\n",
       " ('gmo', 0.1401987075805664),\n",
       " ('oil', 0.12144997715950012),\n",
       " ('silicone', 0.12121882289648056),\n",
       " ('ewg', 0.10684055835008621),\n",
       " ('peanut', 0.10372351109981537)]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tag_model.wv.most_similar('vegan')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('cruelty', 0.29218629002571106),\n",
       " ('free', 0.25725415349006653),\n",
       " ('canadian', 0.23358379304409027),\n",
       " ('talc', 0.22705349326133728),\n",
       " ('organic', 0.21651409566402435),\n",
       " ('water', 0.18804243206977844),\n",
       " ('peanut', 0.17986589670181274),\n",
       " ('usda', 0.17492470145225525),\n",
       " ('non', 0.1721552461385727),\n",
       " ('product', 0.14651453495025635)]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tag_model.wv.most_similar('gmo')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('ewg', 0.240944504737854),\n",
       " ('gmo', 0.21651409566402435),\n",
       " ('trade', 0.17358337342739105),\n",
       " ('hypoallergenic', 0.16875968873500824),\n",
       " ('cruelty', 0.15045318007469177),\n",
       " ('ecocert', 0.13400912284851074),\n",
       " ('free', 0.10918927192687988),\n",
       " ('non', 0.10645356774330139),\n",
       " ('no', 0.09891332685947418),\n",
       " ('oil', 0.09390844404697418)]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tag_model.wv.most_similar('organic')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Reduce word vector dimensionality with PCA and t-SNE\n",
    "\n",
    "We used dimensionality reduction to reduce the number of dimensions from 50 to 3. In particular, we used PCA and t-SNE."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2778"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vocab = list(bigram_model.wv.vocab)\n",
    "len(vocab)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The vector representing each word:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PC\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:1: DeprecationWarning: Call to deprecated `__getitem__` (Method will be removed in 4.0.0, use self.wv.__getitem__() instead).\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "X = bigram_model[vocab]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "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>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>lippie</th>\n",
       "      <td>-0.436240</td>\n",
       "      <td>0.189911</td>\n",
       "      <td>-0.214933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pencil</th>\n",
       "      <td>-2.699123</td>\n",
       "      <td>3.385368</td>\n",
       "      <td>1.178507</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>long_wearing</th>\n",
       "      <td>-1.999066</td>\n",
       "      <td>2.291777</td>\n",
       "      <td>-0.453971</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     x         y         z\n",
       "lippie       -0.436240  0.189911 -0.214933\n",
       "pencil       -2.699123  3.385368  1.178507\n",
       "long_wearing -1.999066  2.291777 -0.453971"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca = PCA(n_components=3, random_state=11, whiten=True)\n",
    "pca_result = pca.fit_transform(X)\n",
    "\n",
    "tmp = pd.DataFrame(pca_result, index=vocab, columns=['x', 'y', 'z'])\n",
    "\n",
    "tmp.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "tmp = tmp.sample(100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(15, 15))\n",
    "ax = fig.add_subplot(111, projection='3d')\n",
    "\n",
    "ax.scatter(tmp['x'], tmp['y'], tmp['z'], alpha = 0.5)\n",
    "\n",
    "for word, row in tmp.iterrows():\n",
    "    x, y, z = row\n",
    "    pos = (x, y, z)\n",
    "    ax.text(x, y, z, s=word, size=8, zorder=1, color='k')\n",
    "    \n",
    "plt.title('PCA')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### TSNE\n",
    "\n",
    "T-SNE is a useful tool to visualize high-dimensional data by dimension reduction while keeping relative pairwise distance between points. T-SNE looking for a new data representation where the neighborhood relations are preserved. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "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>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>lippie</th>\n",
       "      <td>-4.341869</td>\n",
       "      <td>5.552348</td>\n",
       "      <td>0.090714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pencil</th>\n",
       "      <td>-8.331988</td>\n",
       "      <td>17.812258</td>\n",
       "      <td>-8.053832</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>long_wearing</th>\n",
       "      <td>-5.802444</td>\n",
       "      <td>17.370501</td>\n",
       "      <td>-4.650693</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     x          y         z\n",
       "lippie       -4.341869   5.552348  0.090714\n",
       "pencil       -8.331988  17.812258 -8.053832\n",
       "long_wearing -5.802444  17.370501 -4.650693"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tsne = TSNE(n_components=3, random_state=11)\n",
    "tsne_result = tsne.fit_transform(X)\n",
    "\n",
    "tmp = pd.DataFrame(tsne_result, index=vocab, columns=['x', 'y', 'z'])\n",
    "\n",
    "tmp.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "tmp = tmp.sample(100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(15, 15))\n",
    "ax = fig.add_subplot(111, projection='3d')\n",
    "\n",
    "ax.scatter(tmp['x'], tmp['y'], tmp['z'], alpha = 0.5)\n",
    "\n",
    "for word, row in tmp.iterrows():\n",
    "    x, y, z = row\n",
    "    pos = (x, y, z)\n",
    "    ax.text(x, y, z, s=word, size=8, zorder=1, color='k')\n",
    "    \n",
    "plt.title('t-SNE')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}