Association Rules_09-22-2021.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "source": [
    "#Function to generate association rules from frequent itemsets\r\n",
    "from mlxtend.frequent_patterns import association_rules"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "source": [
    "#Import Books dataset for analysis\r\n",
    "import pandas as pd\r\n",
    "import numpy as np\r\n",
    "\r\n",
    "filepath = 'https://raw.githubusercontent.com/jshumway0475/Predictive-Analytics/main/Books.csv'\r\n",
    "books = pd.read_csv(filepath)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "source": [
    "#Review data\r\n",
    "print(books.info())\r\n",
    "print()\r\n",
    "print(books.head())"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1878 entries, 0 to 1877\n",
      "Data columns (total 7 columns):\n",
      " #   Column    Non-Null Count  Dtype \n",
      "---  ------    --------------  ----- \n",
      " 0   ChildBks  1878 non-null   object\n",
      " 1   YouthBks  1878 non-null   object\n",
      " 2   CookBks   1878 non-null   object\n",
      " 3   DoItYBks  1878 non-null   object\n",
      " 4   RefBks    1878 non-null   object\n",
      " 5   ArtBks    1878 non-null   object\n",
      " 6   GeogBks   1878 non-null   object\n",
      "dtypes: object(7)\n",
      "memory usage: 102.8+ KB\n",
      "None\n",
      "\n",
      "  ChildBks YouthBks CookBks DoItYBks RefBks ArtBks GeogBks\n",
      "0      yes       no     yes       no    yes    yes      no\n",
      "1       no      yes     yes      yes    yes    yes     yes\n",
      "2      yes      yes     yes      yes    yes    yes     yes\n",
      "3       no       no      no      yes     no    yes      no\n",
      "4      yes      yes      no      yes    yes    yes      no\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "source": [
    "#Prepare dataframe for modeling\r\n",
    "books = books.replace('no', False)\r\n",
    "books = books.replace('yes', True)\r\n",
    "books.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>ChildBks</th>\n",
       "      <th>YouthBks</th>\n",
       "      <th>CookBks</th>\n",
       "      <th>DoItYBks</th>\n",
       "      <th>RefBks</th>\n",
       "      <th>ArtBks</th>\n",
       "      <th>GeogBks</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ChildBks  YouthBks  CookBks  DoItYBks  RefBks  ArtBks  GeogBks\n",
       "0      True     False     True     False    True    True    False\n",
       "1     False      True     True      True    True    True     True\n",
       "2      True      True     True      True    True    True     True\n",
       "3     False     False    False      True   False    True    False\n",
       "4      True      True    False      True    True    True    False"
      ]
     },
     "metadata": {},
     "execution_count": 5
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "source": [
    "#Create association rules model\r\n",
    "from mlxtend.frequent_patterns import apriori\r\n",
    "from mlxtend.frequent_patterns import association_rules\r\n",
    "\r\n",
    "frequent_itemsets_books = apriori(books, min_support = 0.6, use_colnames = True)\r\n",
    "books_ar = association_rules(frequent_itemsets_books, metric = 'confidence', min_threshold = 0.7)\r\n",
    "books_ar.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>antecedents</th>\n",
       "      <th>consequents</th>\n",
       "      <th>antecedent support</th>\n",
       "      <th>consequent support</th>\n",
       "      <th>support</th>\n",
       "      <th>confidence</th>\n",
       "      <th>lift</th>\n",
       "      <th>leverage</th>\n",
       "      <th>conviction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>(YouthBks)</td>\n",
       "      <td>(RefBks)</td>\n",
       "      <td>0.752929</td>\n",
       "      <td>0.781150</td>\n",
       "      <td>0.630990</td>\n",
       "      <td>0.838048</td>\n",
       "      <td>1.072839</td>\n",
       "      <td>0.042840</td>\n",
       "      <td>1.351326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>(RefBks)</td>\n",
       "      <td>(YouthBks)</td>\n",
       "      <td>0.781150</td>\n",
       "      <td>0.752929</td>\n",
       "      <td>0.630990</td>\n",
       "      <td>0.807771</td>\n",
       "      <td>1.072839</td>\n",
       "      <td>0.042840</td>\n",
       "      <td>1.285297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(YouthBks)</td>\n",
       "      <td>(ArtBks)</td>\n",
       "      <td>0.752929</td>\n",
       "      <td>0.758253</td>\n",
       "      <td>0.612354</td>\n",
       "      <td>0.813296</td>\n",
       "      <td>1.072591</td>\n",
       "      <td>0.041443</td>\n",
       "      <td>1.294809</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>(ArtBks)</td>\n",
       "      <td>(YouthBks)</td>\n",
       "      <td>0.758253</td>\n",
       "      <td>0.752929</td>\n",
       "      <td>0.612354</td>\n",
       "      <td>0.807584</td>\n",
       "      <td>1.072591</td>\n",
       "      <td>0.041443</td>\n",
       "      <td>1.284050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>(YouthBks)</td>\n",
       "      <td>(GeogBks)</td>\n",
       "      <td>0.752929</td>\n",
       "      <td>0.728435</td>\n",
       "      <td>0.600106</td>\n",
       "      <td>0.797030</td>\n",
       "      <td>1.094168</td>\n",
       "      <td>0.051647</td>\n",
       "      <td>1.337957</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  antecedents consequents  antecedent support  consequent support   support  \\\n",
       "0  (YouthBks)    (RefBks)            0.752929            0.781150  0.630990   \n",
       "1    (RefBks)  (YouthBks)            0.781150            0.752929  0.630990   \n",
       "2  (YouthBks)    (ArtBks)            0.752929            0.758253  0.612354   \n",
       "3    (ArtBks)  (YouthBks)            0.758253            0.752929  0.612354   \n",
       "4  (YouthBks)   (GeogBks)            0.752929            0.728435  0.600106   \n",
       "\n",
       "   confidence      lift  leverage  conviction  \n",
       "0    0.838048  1.072839  0.042840    1.351326  \n",
       "1    0.807771  1.072839  0.042840    1.285297  \n",
       "2    0.813296  1.072591  0.041443    1.294809  \n",
       "3    0.807584  1.072591  0.041443    1.284050  \n",
       "4    0.797030  1.094168  0.051647    1.337957  "
      ]
     },
     "metadata": {},
     "execution_count": 6
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "**Sample calculations to verify the output (using row 0, antecedent: RefBks consequent: YouthBks)**  \n",
    "\n",
    "*Number of Antecedant Instances = antecedant support * Count(Transactions)*  \n",
    "1. Number of Antecedant Instances (row 0) = 0.781150 * 1878 = 1467 which is correct!  \n",
    "  \n",
    "*Number of Consequent Instances = consequent support * Count(Transactions)*  \n",
    "1. Number of Consequent Instances (row 0) = 0.752929 * 1878 = 1414 which is correct!  \n",
    "  \n",
    "*Rules Support = Count(Support AUC rules) / Count(Transactions)*  \n",
    "1. Rules Support (row 0) = 0.630990 = Count(Support AUC rules) / 1878  \n",
    "2. Count(Support AUC rules) = 0.630990 * 1878 = 1185 which is correct!  \n",
    "  \n",
    "*Lift = Confidence / Benchmark Confidence*  \n",
    "*Confidence = Count(Support AUC rules) / Number of Antecedant Instances*  \n",
    "*Benchmark Confidence = Number of Consequent Instances / Count(Transactions)*   \n",
    "1. Confidence (row 0) = 1185 / 1467 = 0.807771 which is correct!  \n",
    "2. Benchmark Confidence (row 0) = 1414 / 1878 = 0.752929 which is correct!  \n",
    "3. Lift (row 0) = 0.807771 / 0.752929 = 1.072839 which is correct!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "**Sample calculations to verify the output (using row 1, antecedent: YouthBks consequent: RefBks)**  \n",
    "\n",
    "*Number of Antecedant Instances = antecedant support * Count(Transactions)*  \n",
    "1. Number of Antecedant Instances (row 1) = 0.752929 * 1878 = 1414 which is correct!  \n",
    "  \n",
    "*Number of Consequent Instances = consequent support * Count(Transactions)*  \n",
    "1. Number of Consequent Instances (row 1) = 0.781150 * 1878 = 1467 which is correct!  \n",
    "  \n",
    "*Rules Support = Count(Support AUC rules) / Count(Transactions)*  \n",
    "1. Rules Support (row 1) = 0.630990 = Count(Support AUC rules) / 1878  \n",
    "2. Count(Support AUC rules) = 0.630990 * 1878 = 1185 which is correct!  \n",
    "  \n",
    "*Lift = Confidence / Benchmark Confidence*  \n",
    "*Confidence = Count(Support AUC rules) / Number of Antecedant Instances*  \n",
    "*Benchmark Confidence = Number of Consequent Instances / Count(Transactions)*   \n",
    "1. Confidence (row 1) = 1185 / 1414 = 0.838048 which is correct!  \n",
    "2. Benchmark Confidence (row 1) = 1467 / 1878 = 0.781150 which is correct!  \n",
    "3. Lift (row 1) = 0.838048 / 0.781150 = 1.072839 which is correct!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "source": [
    "#Filter rules to the ArtBks consequent\r\n",
    "fname = 'ArtBks'\r\n",
    "books_ar_filtered = books_ar[books_ar['consequents'].astype(str).str.contains(fname)]\r\n",
    "books_ar_filtered.sort_values('lift', ascending = False)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>antecedents</th>\n",
       "      <th>consequents</th>\n",
       "      <th>antecedent support</th>\n",
       "      <th>consequent support</th>\n",
       "      <th>support</th>\n",
       "      <th>confidence</th>\n",
       "      <th>lift</th>\n",
       "      <th>leverage</th>\n",
       "      <th>conviction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>(GeogBks)</td>\n",
       "      <td>(ArtBks)</td>\n",
       "      <td>0.728435</td>\n",
       "      <td>0.758253</td>\n",
       "      <td>0.613951</td>\n",
       "      <td>0.842836</td>\n",
       "      <td>1.111550</td>\n",
       "      <td>0.061613</td>\n",
       "      <td>1.538183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>(DoItYBks)</td>\n",
       "      <td>(ArtBks)</td>\n",
       "      <td>0.717785</td>\n",
       "      <td>0.758253</td>\n",
       "      <td>0.600639</td>\n",
       "      <td>0.836795</td>\n",
       "      <td>1.103583</td>\n",
       "      <td>0.056376</td>\n",
       "      <td>1.481247</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(YouthBks)</td>\n",
       "      <td>(ArtBks)</td>\n",
       "      <td>0.752929</td>\n",
       "      <td>0.758253</td>\n",
       "      <td>0.612354</td>\n",
       "      <td>0.813296</td>\n",
       "      <td>1.072591</td>\n",
       "      <td>0.041443</td>\n",
       "      <td>1.294809</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>(RefBks)</td>\n",
       "      <td>(ArtBks)</td>\n",
       "      <td>0.781150</td>\n",
       "      <td>0.758253</td>\n",
       "      <td>0.630458</td>\n",
       "      <td>0.807089</td>\n",
       "      <td>1.064406</td>\n",
       "      <td>0.038148</td>\n",
       "      <td>1.253153</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   antecedents consequents  antecedent support  consequent support   support  \\\n",
       "14   (GeogBks)    (ArtBks)            0.728435            0.758253  0.613951   \n",
       "9   (DoItYBks)    (ArtBks)            0.717785            0.758253  0.600639   \n",
       "2   (YouthBks)    (ArtBks)            0.752929            0.758253  0.612354   \n",
       "10    (RefBks)    (ArtBks)            0.781150            0.758253  0.630458   \n",
       "\n",
       "    confidence      lift  leverage  conviction  \n",
       "14    0.842836  1.111550  0.061613    1.538183  \n",
       "9     0.836795  1.103583  0.056376    1.481247  \n",
       "2     0.813296  1.072591  0.041443    1.294809  \n",
       "10    0.807089  1.064406  0.038148    1.253153  "
      ]
     },
     "metadata": {},
     "execution_count": 7
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "source": [
    "#Import StatisticsComCourses dataset for analysis\r\n",
    "filepath2 = 'https://raw.githubusercontent.com/jshumway0475/Predictive-Analytics/main/StatisticsComCourses.csv'\r\n",
    "StatisticsComCourses = pd.read_csv(filepath2)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "source": [
    "#Review data\r\n",
    "print(StatisticsComCourses.info())\r\n",
    "print()\r\n",
    "print(StatisticsComCourses.head())"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 365 entries, 0 to 364\n",
      "Data columns (total 8 columns):\n",
      " #   Column      Non-Null Count  Dtype \n",
      "---  ------      --------------  ----- \n",
      " 0   Intro       365 non-null    object\n",
      " 1   DataMining  365 non-null    object\n",
      " 2   Survey      365 non-null    object\n",
      " 3   Cat Data    365 non-null    object\n",
      " 4   Regression  365 non-null    object\n",
      " 5   Forecast    365 non-null    object\n",
      " 6   DOE         365 non-null    object\n",
      " 7   SW          365 non-null    object\n",
      "dtypes: object(8)\n",
      "memory usage: 22.9+ KB\n",
      "None\n",
      "\n",
      "  Intro DataMining Survey Cat Data Regression Forecast  DOE   SW\n",
      "0    no         no    yes      yes        yes      yes  yes  yes\n",
      "1   yes        yes     no      yes        yes      yes  yes  yes\n",
      "2   yes         no    yes       no         no      yes  yes   no\n",
      "3    no        yes    yes      yes        yes      yes  yes  yes\n",
      "4    no         no    yes      yes        yes      yes  yes  yes\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "source": [
    "#Prepare dataframe for modeling\r\n",
    "StatisticsComCourses = StatisticsComCourses.replace('no', False)\r\n",
    "StatisticsComCourses = StatisticsComCourses.replace('yes', True)\r\n",
    "StatisticsComCourses.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>Intro</th>\n",
       "      <th>DataMining</th>\n",
       "      <th>Survey</th>\n",
       "      <th>Cat Data</th>\n",
       "      <th>Regression</th>\n",
       "      <th>Forecast</th>\n",
       "      <th>DOE</th>\n",
       "      <th>SW</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Intro  DataMining  Survey  Cat Data  Regression  Forecast   DOE     SW\n",
       "0  False       False    True      True        True      True  True   True\n",
       "1   True        True   False      True        True      True  True   True\n",
       "2   True       False    True     False       False      True  True  False\n",
       "3  False        True    True      True        True      True  True   True\n",
       "4  False       False    True      True        True      True  True   True"
      ]
     },
     "metadata": {},
     "execution_count": 10
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "source": [
    "#Create association rules model\r\n",
    "frequent_itemsets_StatisticsComCourses = apriori(StatisticsComCourses, min_support = 0.6, use_colnames = True)\r\n",
    "StatisticsComCourses_ar = association_rules(frequent_itemsets_StatisticsComCourses, \r\n",
    "                                            metric = 'confidence', min_threshold = 0.8)\r\n",
    "StatisticsComCourses_ar.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>antecedents</th>\n",
       "      <th>consequents</th>\n",
       "      <th>antecedent support</th>\n",
       "      <th>consequent support</th>\n",
       "      <th>support</th>\n",
       "      <th>confidence</th>\n",
       "      <th>lift</th>\n",
       "      <th>leverage</th>\n",
       "      <th>conviction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>(Survey)</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.813699</td>\n",
       "      <td>0.665753</td>\n",
       "      <td>0.810000</td>\n",
       "      <td>0.995455</td>\n",
       "      <td>-0.003040</td>\n",
       "      <td>0.980534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>(Survey)</td>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>0.813699</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.665753</td>\n",
       "      <td>0.818182</td>\n",
       "      <td>0.995455</td>\n",
       "      <td>-0.003040</td>\n",
       "      <td>0.979452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(Cat Data)</td>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>0.791781</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.663014</td>\n",
       "      <td>0.837370</td>\n",
       "      <td>1.018800</td>\n",
       "      <td>0.012235</td>\n",
       "      <td>1.095016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>(Cat Data)</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.791781</td>\n",
       "      <td>0.663014</td>\n",
       "      <td>0.806667</td>\n",
       "      <td>1.018800</td>\n",
       "      <td>0.012235</td>\n",
       "      <td>1.076996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>(Regression)</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.791781</td>\n",
       "      <td>0.657534</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>1.010381</td>\n",
       "      <td>0.006755</td>\n",
       "      <td>1.041096</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    antecedents   consequents  antecedent support  consequent support  \\\n",
       "0  (DataMining)      (Survey)            0.821918            0.813699   \n",
       "1      (Survey)  (DataMining)            0.813699            0.821918   \n",
       "2    (Cat Data)  (DataMining)            0.791781            0.821918   \n",
       "3  (DataMining)    (Cat Data)            0.821918            0.791781   \n",
       "4  (DataMining)  (Regression)            0.821918            0.791781   \n",
       "\n",
       "    support  confidence      lift  leverage  conviction  \n",
       "0  0.665753    0.810000  0.995455 -0.003040    0.980534  \n",
       "1  0.665753    0.818182  0.995455 -0.003040    0.979452  \n",
       "2  0.663014    0.837370  1.018800  0.012235    1.095016  \n",
       "3  0.663014    0.806667  1.018800  0.012235    1.076996  \n",
       "4  0.657534    0.800000  1.010381  0.006755    1.041096  "
      ]
     },
     "metadata": {},
     "execution_count": 11
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "**Sample calculations to verify the output (using row 0, antecedent: Survey consequent: DataMining)**  \n",
    "\n",
    "*Number of Antecedant Instances = antecedant support * Count(Transactions)*  \n",
    "1. Number of Antecedant Instances (row 0) = 0.813699 * 365 = 297 which is correct!  \n",
    "  \n",
    "*Number of Consequent Instances = consequent support * Count(Transactions)*  \n",
    "1. Number of Consequent Instances (row 0) = 0.821918 * 365 = 300 which is correct!  \n",
    "  \n",
    "*Rules Support = Count(Support AUC rules) / Count(Transactions)*  \n",
    "1. Rules Support (row 0) = 0.665753\t = Count(Support AUC rules) / 365  \n",
    "2. Count(Support AUC rules) = 0.665753 * 365 = 243 which is correct!  \n",
    "  \n",
    "*Lift = Confidence / Benchmark Confidence*  \n",
    "*Confidence = Count(Support AUC rules) / Number of Antecedant Instances*  \n",
    "*Benchmark Confidence = Number of Consequent Instances / Count(Transactions)*   \n",
    "1. Confidence (row 0) = 243 / 297 = 0.818182 which is correct!  \n",
    "2. Benchmark Confidence (row 0) = 300 / 365 = 0.821918 which is correct!  \n",
    "3. Lift (row 0) = 0.818182 / 0.821918 = 0.995455 which is correct!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "source": [
    "#Filter rules to the DataMining consequent\r\n",
    "fname = 'DataMining'\r\n",
    "StatisticsComCourses_ar_filtered = StatisticsComCourses_ar[StatisticsComCourses_ar['consequents'].astype(str).str.contains(fname)]\r\n",
    "StatisticsComCourses_ar_filtered.sort_values('lift', ascending = False)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>antecedents</th>\n",
       "      <th>consequents</th>\n",
       "      <th>antecedent support</th>\n",
       "      <th>consequent support</th>\n",
       "      <th>support</th>\n",
       "      <th>confidence</th>\n",
       "      <th>lift</th>\n",
       "      <th>leverage</th>\n",
       "      <th>conviction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>(Forecast)</td>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>0.860274</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.723288</td>\n",
       "      <td>0.840764</td>\n",
       "      <td>1.022930</td>\n",
       "      <td>0.016213</td>\n",
       "      <td>1.118356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(Cat Data)</td>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>0.791781</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.663014</td>\n",
       "      <td>0.837370</td>\n",
       "      <td>1.018800</td>\n",
       "      <td>0.012235</td>\n",
       "      <td>1.095016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>(Regression)</td>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>0.791781</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.657534</td>\n",
       "      <td>0.830450</td>\n",
       "      <td>1.010381</td>\n",
       "      <td>0.006755</td>\n",
       "      <td>1.050321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>(SW)</td>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>0.778082</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.638356</td>\n",
       "      <td>0.820423</td>\n",
       "      <td>0.998181</td>\n",
       "      <td>-0.001163</td>\n",
       "      <td>0.991673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>(Survey)</td>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>0.813699</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.665753</td>\n",
       "      <td>0.818182</td>\n",
       "      <td>0.995455</td>\n",
       "      <td>-0.003040</td>\n",
       "      <td>0.979452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>(DOE)</td>\n",
       "      <td>(DataMining)</td>\n",
       "      <td>0.827397</td>\n",
       "      <td>0.821918</td>\n",
       "      <td>0.673973</td>\n",
       "      <td>0.814570</td>\n",
       "      <td>0.991060</td>\n",
       "      <td>-0.006080</td>\n",
       "      <td>0.960372</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     antecedents   consequents  antecedent support  consequent support  \\\n",
       "6     (Forecast)  (DataMining)            0.860274            0.821918   \n",
       "2     (Cat Data)  (DataMining)            0.791781            0.821918   \n",
       "5   (Regression)  (DataMining)            0.791781            0.821918   \n",
       "10          (SW)  (DataMining)            0.778082            0.821918   \n",
       "1       (Survey)  (DataMining)            0.813699            0.821918   \n",
       "8          (DOE)  (DataMining)            0.827397            0.821918   \n",
       "\n",
       "     support  confidence      lift  leverage  conviction  \n",
       "6   0.723288    0.840764  1.022930  0.016213    1.118356  \n",
       "2   0.663014    0.837370  1.018800  0.012235    1.095016  \n",
       "5   0.657534    0.830450  1.010381  0.006755    1.050321  \n",
       "10  0.638356    0.820423  0.998181 -0.001163    0.991673  \n",
       "1   0.665753    0.818182  0.995455 -0.003040    0.979452  \n",
       "8   0.673973    0.814570  0.991060 -0.006080    0.960372  "
      ]
     },
     "metadata": {},
     "execution_count": 12
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [],
   "outputs": [],
   "metadata": {}
  }
 ],
 "metadata": {
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3.8.8 64-bit ('base': conda)"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  },
  "interpreter": {
   "hash": "0d3320a2410d5679c191bc2295f072288b4e5c4c731fe80c4380901918512b12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}