assignment4.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Import the dataset and run the Apriori Rule Learning algorithm. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "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>Class</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3rd</td>\n",
       "      <td>Male</td>\n",
       "      <td>Child</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3rd</td>\n",
       "      <td>Male</td>\n",
       "      <td>Child</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3rd</td>\n",
       "      <td>Male</td>\n",
       "      <td>Child</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3rd</td>\n",
       "      <td>Male</td>\n",
       "      <td>Child</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3rd</td>\n",
       "      <td>Male</td>\n",
       "      <td>Child</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Class   Sex    Age Survived\n",
       "0   3rd  Male  Child       No\n",
       "1   3rd  Male  Child       No\n",
       "2   3rd  Male  Child       No\n",
       "3   3rd  Male  Child       No\n",
       "4   3rd  Male  Child       No"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Assignment 4\n",
    "# Importing the libraries\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "dataset = pd.read_csv('titanic.csv')\n",
    "dataset.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "records = []  \n",
    "for i in range(0, len(dataset)):\n",
    "    records.append([str(dataset.values[i,j]) for j in range(0, len(dataset.columns))])\n",
    "    \n",
    "\n",
    "# Training Apriorion the dataset\n",
    "from apyori import apriori\n",
    "association_rules= apriori(records, min_support=0.01, min_confidence=0.3, min_lift=1, max_length=4)  \n",
    "association_rules= list(association_rules) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "115\n"
     ]
    }
   ],
   "source": [
    "sum = 0\n",
    "for r in range(0, len(association_rules)):\n",
    "    sum += len(association_rules[r][2])\n",
    "print(sum)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display the rules you get in a proper format where you first show the rule:\n",
    "{A}{B} followed by Support, Confidence, Lift below that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n",
      "Rule: [] -> ['3rd']\n",
      "Support: 0.32076328941390275\n",
      "Confidence: 0.32076328941390275\n",
      "Lift: 1.0\n",
      "=====================================\n",
      "2\n",
      "Rule: [] -> ['Adult']\n",
      "Support: 0.9504770558836892\n",
      "Confidence: 0.9504770558836892\n",
      "Lift: 1.0\n",
      "=====================================\n",
      "3\n",
      "Rule: [] -> ['Crew']\n",
      "Support: 0.4020899591094957\n",
      "Confidence: 0.4020899591094957\n",
      "Lift: 1.0\n",
      "=====================================\n",
      "4\n",
      "Rule: [] -> ['Male']\n",
      "Support: 0.7864606996819627\n",
      "Confidence: 0.7864606996819627\n",
      "Lift: 1.0\n",
      "=====================================\n",
      "5\n",
      "Rule: [] -> ['No']\n",
      "Support: 0.6769650159018628\n",
      "Confidence: 0.6769650159018628\n",
      "Lift: 1.0\n",
      "=====================================\n",
      "6\n",
      "Rule: [] -> ['Yes']\n",
      "Support: 0.3230349840981372\n",
      "Confidence: 0.3230349840981372\n",
      "Lift: 1.0\n",
      "=====================================\n",
      "7\n",
      "Rule: ['1st'] -> ['Adult']\n",
      "Support: 0.1449341208541572\n",
      "Confidence: 0.9815384615384615\n",
      "Lift: 1.032679805853802\n",
      "=====================================\n",
      "8\n",
      "Rule: ['1st'] -> ['Female']\n",
      "Support: 0.06587914584279873\n",
      "Confidence: 0.4461538461538462\n",
      "Lift: 2.089328968903437\n",
      "=====================================\n",
      "9\n",
      "Rule: ['Female'] -> ['1st']\n",
      "Support: 0.06587914584279873\n",
      "Confidence: 0.3085106382978724\n",
      "Lift: 2.089328968903437\n",
      "=====================================\n",
      "10\n",
      "Rule: ['1st'] -> ['Yes']\n",
      "Support: 0.09223080417991822\n",
      "Confidence: 0.6246153846153846\n",
      "Lift: 1.933584334090663\n",
      "=====================================\n",
      "11\n",
      "Rule: ['2nd'] -> ['Female']\n",
      "Support: 0.0481599273057701\n",
      "Confidence: 0.37192982456140344\n",
      "Lift: 1.7417394550205298\n",
      "=====================================\n",
      "12\n",
      "Rule: ['2nd'] -> ['Yes']\n",
      "Support: 0.05361199454793276\n",
      "Confidence: 0.4140350877192982\n",
      "Lift: 1.281703555654255\n",
      "=====================================\n",
      "13\n",
      "Rule: ['Child'] -> ['3rd']\n",
      "Support: 0.03589277601090413\n",
      "Confidence: 0.7247706422018348\n",
      "Lift: 2.2595186734932557\n",
      "=====================================\n",
      "14\n",
      "Rule: ['Female'] -> ['3rd']\n",
      "Support: 0.08905043162199\n",
      "Confidence: 0.41702127659574467\n",
      "Lift: 1.3000904104635043\n",
      "=====================================\n",
      "15\n",
      "Rule: ['3rd'] -> ['No']\n",
      "Support: 0.23989095865515675\n",
      "Confidence: 0.7478753541076488\n",
      "Lift: 1.104747419054319\n",
      "=====================================\n",
      "16\n",
      "Rule: ['No'] -> ['3rd']\n",
      "Support: 0.23989095865515675\n",
      "Confidence: 0.35436241610738256\n",
      "Lift: 1.1047474190543187\n",
      "=====================================\n",
      "17\n",
      "Rule: ['Adult'] -> ['Crew']\n",
      "Support: 0.4020899591094957\n",
      "Confidence: 0.42304015296367115\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "18\n",
      "Rule: ['Crew'] -> ['Adult']\n",
      "Support: 0.4020899591094957\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "19\n",
      "Rule: ['Adult'] -> ['Male']\n",
      "Support: 0.7573830077237619\n",
      "Confidence: 0.7968451242829827\n",
      "Lift: 1.0132039968497082\n",
      "=====================================\n",
      "20\n",
      "Rule: ['Male'] -> ['Adult']\n",
      "Support: 0.7573830077237619\n",
      "Confidence: 0.9630271519352974\n",
      "Lift: 1.0132039968497082\n",
      "=====================================\n",
      "21\n",
      "Rule: ['Adult'] -> ['No']\n",
      "Support: 0.6533393911858246\n",
      "Confidence: 0.6873804971319312\n",
      "Lift: 1.015385553145893\n",
      "=====================================\n",
      "22\n",
      "Rule: ['No'] -> ['Adult']\n",
      "Support: 0.6533393911858246\n",
      "Confidence: 0.9651006711409397\n",
      "Lift: 1.015385553145893\n",
      "=====================================\n",
      "23\n",
      "Rule: ['Child'] -> ['Female']\n",
      "Support: 0.02044525215810995\n",
      "Confidence: 0.41284403669724773\n",
      "Lift: 1.9333398399375368\n",
      "=====================================\n",
      "24\n",
      "Rule: ['Child'] -> ['Yes']\n",
      "Support: 0.025897319400272602\n",
      "Confidence: 0.5229357798165137\n",
      "Lift: 1.6188208880114583\n",
      "=====================================\n",
      "25\n",
      "Rule: ['Crew'] -> ['Male']\n",
      "Support: 0.39164016356201725\n",
      "Confidence: 0.9740112994350282\n",
      "Lift: 1.2384742172481207\n",
      "=====================================\n",
      "26\n",
      "Rule: ['Male'] -> ['Crew']\n",
      "Support: 0.39164016356201725\n",
      "Confidence: 0.49797804737146156\n",
      "Lift: 1.2384742172481207\n",
      "=====================================\n",
      "27\n",
      "Rule: ['Crew'] -> ['No']\n",
      "Support: 0.3057701044979555\n",
      "Confidence: 0.76045197740113\n",
      "Lift: 1.1233253706442197\n",
      "=====================================\n",
      "28\n",
      "Rule: ['No'] -> ['Crew']\n",
      "Support: 0.3057701044979555\n",
      "Confidence: 0.4516778523489933\n",
      "Lift: 1.1233253706442194\n",
      "=====================================\n",
      "29\n",
      "Rule: ['Female'] -> ['Yes']\n",
      "Support: 0.1562925942753294\n",
      "Confidence: 0.7319148936170213\n",
      "Lift: 2.2657449800999494\n",
      "=====================================\n",
      "30\n",
      "Rule: ['Yes'] -> ['Female']\n",
      "Support: 0.1562925942753294\n",
      "Confidence: 0.48382559774964845\n",
      "Lift: 2.2657449800999494\n",
      "=====================================\n",
      "31\n",
      "Rule: ['Male'] -> ['No']\n",
      "Support: 0.6197183098591549\n",
      "Confidence: 0.7879838243789716\n",
      "Lift: 1.163994897622897\n",
      "=====================================\n",
      "32\n",
      "Rule: ['No'] -> ['Male']\n",
      "Support: 0.6197183098591549\n",
      "Confidence: 0.9154362416107382\n",
      "Lift: 1.163994897622897\n",
      "=====================================\n",
      "33\n",
      "Rule: ['1st', 'Adult'] -> ['Female']\n",
      "Support: 0.06542480690595184\n",
      "Confidence: 0.4514106583072101\n",
      "Lift: 2.1139465083705735\n",
      "=====================================\n",
      "34\n",
      "Rule: ['1st', 'Female'] -> ['Adult']\n",
      "Support: 0.06542480690595184\n",
      "Confidence: 0.993103448275862\n",
      "Lift: 1.0448473659919562\n",
      "=====================================\n",
      "35\n",
      "Rule: ['Female', 'Adult'] -> ['1st']\n",
      "Support: 0.06542480690595184\n",
      "Confidence: 0.3388235294117647\n",
      "Lift: 2.2946171945701357\n",
      "=====================================\n",
      "36\n",
      "Rule: ['Male', '1st'] -> ['Adult']\n",
      "Support: 0.07950931394820536\n",
      "Confidence: 0.9722222222222222\n",
      "Lift: 1.0228781601869557\n",
      "=====================================\n",
      "37\n",
      "Rule: ['1st', 'No'] -> ['Adult']\n",
      "Support: 0.05542935029532031\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "38\n",
      "Rule: ['1st', 'Adult'] -> ['Yes']\n",
      "Support: 0.08950477055883689\n",
      "Confidence: 0.6175548589341693\n",
      "Lift: 1.9117274887680826\n",
      "=====================================\n",
      "39\n",
      "Rule: ['1st', 'Yes'] -> ['Adult']\n",
      "Support: 0.08950477055883689\n",
      "Confidence: 0.9704433497536946\n",
      "Lift: 1.021006602680632\n",
      "=====================================\n",
      "40\n",
      "Rule: ['Yes', 'Adult'] -> ['1st']\n",
      "Support: 0.08950477055883689\n",
      "Confidence: 0.301223241590214\n",
      "Lift: 2.0399764761232646\n",
      "=====================================\n",
      "41\n",
      "Rule: ['1st', 'Female'] -> ['Yes']\n",
      "Support: 0.06406179009541117\n",
      "Confidence: 0.9724137931034481\n",
      "Lift: 3.0102429797759345\n",
      "=====================================\n",
      "42\n",
      "Rule: ['1st', 'Yes'] -> ['Female']\n",
      "Support: 0.06406179009541117\n",
      "Confidence: 0.6945812807881773\n",
      "Lift: 3.252709359605911\n",
      "=====================================\n",
      "43\n",
      "Rule: ['Female', 'Yes'] -> ['1st']\n",
      "Support: 0.06406179009541117\n",
      "Confidence: 0.4098837209302325\n",
      "Lift: 2.775858676207513\n",
      "=====================================\n",
      "44\n",
      "Rule: ['1st', 'No'] -> ['Male']\n",
      "Support: 0.05361199454793276\n",
      "Confidence: 0.9672131147540983\n",
      "Lift: 1.2298301938612193\n",
      "=====================================\n",
      "45\n",
      "Rule: ['Male', '1st'] -> ['Yes']\n",
      "Support: 0.028169014084507043\n",
      "Confidence: 0.34444444444444444\n",
      "Lift: 1.0662759806219722\n",
      "=====================================\n",
      "46\n",
      "Rule: ['Adult', '2nd'] -> ['Female']\n",
      "Support: 0.04225352112676056\n",
      "Confidence: 0.3563218390804598\n",
      "Lift: 1.668647591098068\n",
      "=====================================\n",
      "47\n",
      "Rule: ['No', '2nd'] -> ['Adult']\n",
      "Support: 0.07587460245343026\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "48\n",
      "Rule: ['Adult', '2nd'] -> ['Yes']\n",
      "Support: 0.04270786006360745\n",
      "Confidence: 0.3601532567049809\n",
      "Lift: 1.1149048073244205\n",
      "=====================================\n",
      "49\n",
      "Rule: ['Child', '2nd'] -> ['Yes']\n",
      "Support: 0.010904134484325307\n",
      "Confidence: 1.0\n",
      "Lift: 3.09563994374121\n",
      "=====================================\n",
      "50\n",
      "Rule: ['Yes', 'Child'] -> ['2nd']\n",
      "Support: 0.010904134484325307\n",
      "Confidence: 0.4210526315789474\n",
      "Lift: 3.251708217913204\n",
      "=====================================\n",
      "51\n",
      "Rule: ['Female', '2nd'] -> ['Yes']\n",
      "Support: 0.04225352112676056\n",
      "Confidence: 0.8773584905660378\n",
      "Lift: 2.7159859883767217\n",
      "=====================================\n",
      "52\n",
      "Rule: ['Yes', '2nd'] -> ['Female']\n",
      "Support: 0.04225352112676056\n",
      "Confidence: 0.788135593220339\n",
      "Lift: 3.690822214208439\n",
      "=====================================\n",
      "53\n",
      "Rule: ['Male', '2nd'] -> ['No']\n",
      "Support: 0.06996819627442072\n",
      "Confidence: 0.8603351955307263\n",
      "Lift: 1.270870983465187\n",
      "=====================================\n",
      "54\n",
      "Rule: ['No', '2nd'] -> ['Male']\n",
      "Support: 0.06996819627442072\n",
      "Confidence: 0.9221556886227544\n",
      "Lift: 1.1725388045399667\n",
      "=====================================\n",
      "55\n",
      "Rule: ['Female', 'Adult'] -> ['3rd']\n",
      "Support: 0.07496592457973648\n",
      "Confidence: 0.38823529411764707\n",
      "Lift: 1.2103482752874521\n",
      "=====================================\n",
      "56\n",
      "Rule: ['3rd', 'Adult'] -> ['No']\n",
      "Support: 0.2162653339391186\n",
      "Confidence: 0.759170653907496\n",
      "Lift: 1.121432623657986\n",
      "=====================================\n",
      "57\n",
      "Rule: ['No', 'Adult'] -> ['3rd']\n",
      "Support: 0.2162653339391186\n",
      "Confidence: 0.3310152990264256\n",
      "Lift: 1.0319612934237434\n",
      "=====================================\n",
      "58\n",
      "Rule: ['3rd', 'Child'] -> ['Female']\n",
      "Support: 0.014084507042253521\n",
      "Confidence: 0.3924050632911393\n",
      "Lift: 1.8376245623485055\n",
      "=====================================\n",
      "59\n",
      "Rule: ['Female', 'Child'] -> ['3rd']\n",
      "Support: 0.014084507042253521\n",
      "Confidence: 0.6888888888888889\n",
      "Lift: 2.147655020459553\n",
      "=====================================\n",
      "60\n",
      "Rule: ['Male', 'Child'] -> ['3rd']\n",
      "Support: 0.021808268968650613\n",
      "Confidence: 0.75\n",
      "Lift: 2.338172804532578\n",
      "=====================================\n",
      "61\n",
      "Rule: ['No', 'Child'] -> ['3rd']\n",
      "Support: 0.023625624716038164\n",
      "Confidence: 1.0\n",
      "Lift: 3.1175637393767706\n",
      "=====================================\n",
      "62\n",
      "Rule: ['3rd', 'Child'] -> ['Yes']\n",
      "Support: 0.01226715129486597\n",
      "Confidence: 0.3417721518987342\n",
      "Lift: 1.0580035250761097\n",
      "=====================================\n",
      "63\n",
      "Rule: ['Yes', 'Child'] -> ['3rd']\n",
      "Support: 0.01226715129486597\n",
      "Confidence: 0.4736842105263158\n",
      "Lift: 1.4767407186521546\n",
      "=====================================\n",
      "64\n",
      "Rule: ['Female', 'No'] -> ['3rd']\n",
      "Support: 0.0481599273057701\n",
      "Confidence: 0.8412698412698413\n",
      "Lift: 2.622712352174109\n",
      "=====================================\n",
      "65\n",
      "Rule: ['Female', '3rd'] -> ['Yes']\n",
      "Support: 0.0408905043162199\n",
      "Confidence: 0.45918367346938777\n",
      "Lift: 1.4214673211056577\n",
      "=====================================\n",
      "66\n",
      "Rule: ['Yes', '3rd'] -> ['Female']\n",
      "Support: 0.0408905043162199\n",
      "Confidence: 0.5056179775280898\n",
      "Lift: 2.3677982309347354\n",
      "=====================================\n",
      "67\n",
      "Rule: ['Male', '3rd'] -> ['No']\n",
      "Support: 0.19173103134938665\n",
      "Confidence: 0.8274509803921569\n",
      "Lift: 1.2222950388208977\n",
      "=====================================\n",
      "68\n",
      "Rule: ['3rd', 'No'] -> ['Male']\n",
      "Support: 0.19173103134938665\n",
      "Confidence: 0.7992424242424243\n",
      "Lift: 1.0162522101430247\n",
      "=====================================\n",
      "69\n",
      "Rule: ['Female', 'Crew'] -> ['Adult']\n",
      "Support: 0.010449795547478418\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "70\n",
      "Rule: ['Crew', 'Adult'] -> ['Male']\n",
      "Support: 0.39164016356201725\n",
      "Confidence: 0.9740112994350282\n",
      "Lift: 1.2384742172481207\n",
      "=====================================\n",
      "71\n",
      "Rule: ['Male', 'Adult'] -> ['Crew']\n",
      "Support: 0.39164016356201725\n",
      "Confidence: 0.5170965806838632\n",
      "Lift: 1.286022117610376\n",
      "=====================================\n",
      "72\n",
      "Rule: ['Male', 'Crew'] -> ['Adult']\n",
      "Support: 0.39164016356201725\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "73\n",
      "Rule: ['Crew', 'Adult'] -> ['No']\n",
      "Support: 0.3057701044979555\n",
      "Confidence: 0.76045197740113\n",
      "Lift: 1.1233253706442197\n",
      "=====================================\n",
      "74\n",
      "Rule: ['No', 'Adult'] -> ['Crew']\n",
      "Support: 0.3057701044979555\n",
      "Confidence: 0.4680111265646732\n",
      "Lift: 1.1639463158970007\n",
      "=====================================\n",
      "75\n",
      "Rule: ['No', 'Crew'] -> ['Adult']\n",
      "Support: 0.3057701044979555\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "76\n",
      "Rule: ['Yes', 'Crew'] -> ['Adult']\n",
      "Support: 0.0963198546115402\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "77\n",
      "Rule: ['Female', 'Adult'] -> ['Yes']\n",
      "Support: 0.14357110404361653\n",
      "Confidence: 0.7435294117647059\n",
      "Lift: 2.301699346405229\n",
      "=====================================\n",
      "78\n",
      "Rule: ['Yes', 'Adult'] -> ['Female']\n",
      "Support: 0.14357110404361653\n",
      "Confidence: 0.4831804281345565\n",
      "Lift: 2.2627236645194873\n",
      "=====================================\n",
      "79\n",
      "Rule: ['Male', 'Adult'] -> ['No']\n",
      "Support: 0.6038164470695139\n",
      "Confidence: 0.7972405518896222\n",
      "Lift: 1.1776687615497037\n",
      "=====================================\n",
      "80\n",
      "Rule: ['No', 'Adult'] -> ['Male']\n",
      "Support: 0.6038164470695139\n",
      "Confidence: 0.9242002781641169\n",
      "Lift: 1.1751385397106997\n",
      "=====================================\n",
      "81\n",
      "Rule: ['Male', 'No'] -> ['Adult']\n",
      "Support: 0.6038164470695139\n",
      "Confidence: 0.9743401759530793\n",
      "Lift: 1.0251064661915523\n",
      "=====================================\n",
      "82\n",
      "Rule: ['Female', 'Child'] -> ['Yes']\n",
      "Support: 0.012721490231712857\n",
      "Confidence: 0.6222222222222222\n",
      "Lift: 1.9261759649945305\n",
      "=====================================\n",
      "83\n",
      "Rule: ['Yes', 'Child'] -> ['Female']\n",
      "Support: 0.012721490231712857\n",
      "Confidence: 0.4912280701754386\n",
      "Lift: 2.3004106009705114\n",
      "=====================================\n",
      "84\n",
      "Rule: ['Male', 'Child'] -> ['Yes']\n",
      "Support: 0.013175829168559746\n",
      "Confidence: 0.45312500000000006\n",
      "Lift: 1.4027118495077358\n",
      "=====================================\n",
      "85\n",
      "Rule: ['Male', 'Crew'] -> ['No']\n",
      "Support: 0.3044070876874148\n",
      "Confidence: 0.777262180974478\n",
      "Lift: 1.1481570874663263\n",
      "=====================================\n",
      "86\n",
      "Rule: ['No', 'Crew'] -> ['Male']\n",
      "Support: 0.3044070876874148\n",
      "Confidence: 0.9955423476968795\n",
      "Lift: 1.2658513618029068\n",
      "=====================================\n",
      "87\n",
      "Rule: ['Male', 'No'] -> ['Crew']\n",
      "Support: 0.3044070876874148\n",
      "Confidence: 0.4912023460410558\n",
      "Lift: 1.2216230097586032\n",
      "=====================================\n",
      "88\n",
      "Rule: ['Yes', 'Crew'] -> ['Male']\n",
      "Support: 0.08723307587460245\n",
      "Confidence: 0.9056603773584906\n",
      "Lift: 1.1515646970341062\n",
      "=====================================\n",
      "89\n",
      "Rule: ['Male', 'Yes'] -> ['Crew']\n",
      "Support: 0.08723307587460245\n",
      "Confidence: 0.5231607629427792\n",
      "Lift: 1.301103773149217\n",
      "=====================================\n",
      "90\n",
      "Rule: ['1st', 'Female', 'Adult'] -> ['Yes']\n",
      "Support: 0.06360745115856428\n",
      "Confidence: 0.9722222222222221\n",
      "Lift: 3.0096499453039534\n",
      "=====================================\n",
      "91\n",
      "Rule: ['1st', 'Yes', 'Adult'] -> ['Female']\n",
      "Support: 0.06360745115856428\n",
      "Confidence: 0.7106598984771573\n",
      "Lift: 3.3280051841451557\n",
      "=====================================\n",
      "92\n",
      "Rule: ['1st', 'Yes', 'Female'] -> ['Adult']\n",
      "Support: 0.06360745115856428\n",
      "Confidence: 0.9929078014184397\n",
      "Lift: 1.0446415252973162\n",
      "=====================================\n",
      "93\n",
      "Rule: ['Female', 'Yes', 'Adult'] -> ['1st']\n",
      "Support: 0.06360745115856428\n",
      "Confidence: 0.4430379746835443\n",
      "Lift: 3.0003894839337875\n",
      "=====================================\n",
      "94\n",
      "Rule: ['1st', 'No', 'Adult'] -> ['Male']\n",
      "Support: 0.05361199454793276\n",
      "Confidence: 0.9672131147540983\n",
      "Lift: 1.2298301938612193\n",
      "=====================================\n",
      "95\n",
      "Rule: ['Male', '1st', 'No'] -> ['Adult']\n",
      "Support: 0.05361199454793276\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "96\n",
      "Rule: ['Male', '1st', 'Adult'] -> ['Yes']\n",
      "Support: 0.025897319400272602\n",
      "Confidence: 0.3257142857142857\n",
      "Lift: 1.0082941531042797\n",
      "=====================================\n",
      "97\n",
      "Rule: ['Female', 'Adult', '2nd'] -> ['Yes']\n",
      "Support: 0.03634711494775102\n",
      "Confidence: 0.8602150537634409\n",
      "Lift: 2.6629160806375998\n",
      "=====================================\n",
      "98\n",
      "Rule: ['Yes', 'Adult', '2nd'] -> ['Female']\n",
      "Support: 0.03634711494775102\n",
      "Confidence: 0.851063829787234\n",
      "Lift: 3.9855138071525578\n",
      "=====================================\n",
      "99\n",
      "Rule: ['Male', 'Adult', '2nd'] -> ['No']\n",
      "Support: 0.06996819627442072\n",
      "Confidence: 0.9166666666666667\n",
      "Lift: 1.3540827740492172\n",
      "=====================================\n",
      "100\n",
      "Rule: ['Adult', 'No', '2nd'] -> ['Male']\n",
      "Support: 0.06996819627442072\n",
      "Confidence: 0.9221556886227544\n",
      "Lift: 1.1725388045399667\n",
      "=====================================\n",
      "101\n",
      "Rule: ['Male', 'No', '2nd'] -> ['Adult']\n",
      "Support: 0.06996819627442072\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "102\n",
      "Rule: ['Female', 'No', 'Adult'] -> ['3rd']\n",
      "Support: 0.040436165379373015\n",
      "Confidence: 0.81651376146789\n",
      "Lift: 2.5455336954544276\n",
      "=====================================\n",
      "103\n",
      "Rule: ['Female', '3rd', 'Adult'] -> ['Yes']\n",
      "Support: 0.03452975920036347\n",
      "Confidence: 0.4606060606060606\n",
      "Lift: 1.4258705195414056\n",
      "=====================================\n",
      "104\n",
      "Rule: ['Yes', '3rd', 'Adult'] -> ['Female']\n",
      "Support: 0.03452975920036347\n",
      "Confidence: 0.5033112582781457\n",
      "Lift: 2.3569959137663803\n",
      "=====================================\n",
      "105\n",
      "Rule: ['Male', '3rd', 'Adult'] -> ['No']\n",
      "Support: 0.17582916855974556\n",
      "Confidence: 0.8376623376623377\n",
      "Lift: 1.237379063889131\n",
      "=====================================\n",
      "106\n",
      "Rule: ['3rd', 'No', 'Adult'] -> ['Male']\n",
      "Support: 0.17582916855974556\n",
      "Confidence: 0.8130252100840335\n",
      "Lift: 1.0337772890785428\n",
      "=====================================\n",
      "107\n",
      "Rule: ['Male', '3rd', 'Child'] -> ['No']\n",
      "Support: 0.01590186278964107\n",
      "Confidence: 0.7291666666666666\n",
      "Lift: 1.07711129753915\n",
      "=====================================\n",
      "108\n",
      "Rule: ['Male', 'No', 'Child'] -> ['3rd']\n",
      "Support: 0.01590186278964107\n",
      "Confidence: 1.0\n",
      "Lift: 3.1175637393767706\n",
      "=====================================\n",
      "109\n",
      "Rule: ['Male', 'Crew', 'Adult'] -> ['No']\n",
      "Support: 0.3044070876874148\n",
      "Confidence: 0.777262180974478\n",
      "Lift: 1.1481570874663263\n",
      "=====================================\n",
      "110\n",
      "Rule: ['No', 'Crew', 'Adult'] -> ['Male']\n",
      "Support: 0.3044070876874148\n",
      "Confidence: 0.9955423476968795\n",
      "Lift: 1.2658513618029068\n",
      "=====================================\n",
      "111\n",
      "Rule: ['Male', 'No', 'Adult'] -> ['Crew']\n",
      "Support: 0.3044070876874148\n",
      "Confidence: 0.5041384499623777\n",
      "Lift: 1.2537951732962636\n",
      "=====================================\n",
      "112\n",
      "Rule: ['Male', 'Crew', 'No'] -> ['Adult']\n",
      "Support: 0.3044070876874148\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n",
      "113\n",
      "Rule: ['Yes', 'Crew', 'Adult'] -> ['Male']\n",
      "Support: 0.08723307587460245\n",
      "Confidence: 0.9056603773584906\n",
      "Lift: 1.1515646970341062\n",
      "=====================================\n",
      "114\n",
      "Rule: ['Male', 'Yes', 'Adult'] -> ['Crew']\n",
      "Support: 0.08723307587460245\n",
      "Confidence: 0.5680473372781065\n",
      "Lift: 1.4127369371176413\n",
      "=====================================\n",
      "115\n",
      "Rule: ['Male', 'Yes', 'Crew'] -> ['Adult']\n",
      "Support: 0.08723307587460245\n",
      "Confidence: 1.0\n",
      "Lift: 1.0521032504780115\n",
      "=====================================\n"
     ]
    }
   ],
   "source": [
    "rule_seq = 0\n",
    "for item in association_rules:\n",
    "    for j in range(0, len(item[2])):\n",
    "        a = item[2][j][0]\n",
    "        b = item[2][j][1]\n",
    "        rule_seq = rule_seq + 1\n",
    "        print(rule_seq)\n",
    "        print(\"Rule: \" + str(list(a)) + \" -> \" + str(list(b)))\n",
    "        #second index of the inner list\n",
    "        print(\"Support: \" + str(item[1]))\n",
    "        #third index of the list located at 0th\n",
    "        #of the third index of the inner list\n",
    "        print(\"Confidence: \" + str(item[2][j][2]))\n",
    "        print(\"Lift: \" + str(item[2][j][3]))\n",
    "        print(\"=====================================\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pruning the rules based on consequents: Write a function that takes as parameters: support, confidence, and a consequent attribute.\n",
    "o e.g. If the consequent attribute parameter passed is “Sex”, then your function should only display rules where the consequent has “Sex” only.\n",
    "▪ If Survived=Yes and Class=3rd  Sex=Male (should be displayed)\n",
    "▪ If Survived=Yes and Class=3rd  Sex=Male and Age=Adult (should not be displayed)\n",
    "o e.g. If the consequent attribute parameter passed is “Sex” and “Age”, then your function should only display rules where the consequent has “Sex” and “Age”.\n",
    "▪ If Survived=Yes and Class=3rd  Sex=Male (should not be displayed)\n",
    "▪ If Survived=Yes and Class=3rd  Sex=Male and Age=Adult (should be displayed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prune_rules_on_consequents(support, confidence, consequents):\n",
    "    association_rules= apriori(records, min_support=support, min_confidence=confidence, min_lift=1, max_length=4)  \n",
    "    association_rules= list(association_rules)\n",
    "\n",
    "    for item in association_rules:\n",
    "        for j in range(0, len(item[2])):\n",
    "            a = item[2][j][0]\n",
    "            b = item[2][j][1]\n",
    "            if(consequents == list(b)):\n",
    "                print(\"Rule: \" + str(list(a)) + \" -> \" + str(list(b)))\n",
    "                #second index of the inner list\n",
    "                print(\"Support: \" + str(item[1]))\n",
    "                #third index of the list located at 0th\n",
    "                #of the third index of the inner list\n",
    "                print(\"Confidence: \" + str(item[2][j][2]))\n",
    "                print(\"Lift: \" + str(item[2][j][3]))\n",
    "                print(\"=====================================\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rule: ['1st'] -> ['Female']\n",
      "Support: 0.06587914584279873\n",
      "Confidence: 0.4461538461538462\n",
      "Lift: 2.089328968903437\n",
      "=====================================\n",
      "Rule: ['2nd'] -> ['Female']\n",
      "Support: 0.0481599273057701\n",
      "Confidence: 0.37192982456140344\n",
      "Lift: 1.7417394550205298\n",
      "=====================================\n",
      "Rule: ['Child'] -> ['Female']\n",
      "Support: 0.02044525215810995\n",
      "Confidence: 0.41284403669724773\n",
      "Lift: 1.9333398399375368\n",
      "=====================================\n",
      "Rule: ['Yes'] -> ['Female']\n",
      "Support: 0.1562925942753294\n",
      "Confidence: 0.48382559774964845\n",
      "Lift: 2.2657449800999494\n",
      "=====================================\n",
      "Rule: ['1st', 'Adult'] -> ['Female']\n",
      "Support: 0.06542480690595184\n",
      "Confidence: 0.4514106583072101\n",
      "Lift: 2.1139465083705735\n",
      "=====================================\n",
      "Rule: ['1st', 'Yes'] -> ['Female']\n",
      "Support: 0.06406179009541117\n",
      "Confidence: 0.6945812807881773\n",
      "Lift: 3.252709359605911\n",
      "=====================================\n",
      "Rule: ['Adult', '2nd'] -> ['Female']\n",
      "Support: 0.04225352112676056\n",
      "Confidence: 0.3563218390804598\n",
      "Lift: 1.668647591098068\n",
      "=====================================\n",
      "Rule: ['Yes', '2nd'] -> ['Female']\n",
      "Support: 0.04225352112676056\n",
      "Confidence: 0.788135593220339\n",
      "Lift: 3.690822214208439\n",
      "=====================================\n",
      "Rule: ['3rd', 'Child'] -> ['Female']\n",
      "Support: 0.014084507042253521\n",
      "Confidence: 0.3924050632911393\n",
      "Lift: 1.8376245623485055\n",
      "=====================================\n",
      "Rule: ['Yes', '3rd'] -> ['Female']\n",
      "Support: 0.0408905043162199\n",
      "Confidence: 0.5056179775280898\n",
      "Lift: 2.3677982309347354\n",
      "=====================================\n",
      "Rule: ['Yes', 'Adult'] -> ['Female']\n",
      "Support: 0.14357110404361653\n",
      "Confidence: 0.4831804281345565\n",
      "Lift: 2.2627236645194873\n",
      "=====================================\n",
      "Rule: ['Yes', 'Child'] -> ['Female']\n",
      "Support: 0.012721490231712857\n",
      "Confidence: 0.4912280701754386\n",
      "Lift: 2.3004106009705114\n",
      "=====================================\n",
      "Rule: ['1st', 'Yes', 'Adult'] -> ['Female']\n",
      "Support: 0.06360745115856428\n",
      "Confidence: 0.7106598984771573\n",
      "Lift: 3.3280051841451557\n",
      "=====================================\n",
      "Rule: ['Yes', 'Adult', '2nd'] -> ['Female']\n",
      "Support: 0.03634711494775102\n",
      "Confidence: 0.851063829787234\n",
      "Lift: 3.9855138071525578\n",
      "=====================================\n",
      "Rule: ['Yes', '3rd', 'Adult'] -> ['Female']\n",
      "Support: 0.03452975920036347\n",
      "Confidence: 0.5033112582781457\n",
      "Lift: 2.3569959137663803\n",
      "=====================================\n"
     ]
    }
   ],
   "source": [
    "prune_rules_on_consequents(0.01,0.3,['Female'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pruning redundant rules: Write a function that takes support and confidence as parameters and prunes redundant rules while displaying the rules.\n",
    "o e.g. Rule 1: If Class=2nd and Age=Child  Survived = Yes\n",
    "Rule 2: If Class=2nd and Age=Child and Sex=Female  Survived = Yes (redundant)\n",
    "Rule 2 does not give us any additional information and hence is a redundant rule. Your function should only display non-redundant rules."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prune_redundant_rules(support, confidence):\n",
    "    association_rules= apriori(records, min_support=support, min_confidence=confidence, min_lift=1, max_length=4)  \n",
    "    association_rules= list(association_rules)\n",
    "    \n",
    "    rules_result = {}\n",
    "    for item in association_rules:\n",
    "        for j in range(0, len(item[2])):\n",
    "            a = item[2][j][0]\n",
    "            b = item[2][j][1]\n",
    "            if str(list(b)) in rules_result:\n",
    "                #if a is aubset of any of the antecedents of b\n",
    "                #replace that antecedent with a, otherwise add a \n",
    "                current_antecedents = rules_result.get(str(list(b)))\n",
    "                new_antecedents = list(current_antecedents)\n",
    "                \n",
    "                \"\"\"\n",
    "                x = list(b)\n",
    "                if (x[0] == 'Female'):\n",
    "                    print(\"PRE  \", str(list(b)), \"\\t\", new_antecedents, \"\\t\", str(list(a)))\n",
    "                \"\"\"\n",
    "                \n",
    "                setNew = set(list(a))\n",
    "                isNewSuperSet = False\n",
    "                \n",
    "                for current_antecedent in current_antecedents:\n",
    "                    setExisting = set(current_antecedent)\n",
    "                    \n",
    "                    if (setNew.issuperset(setExisting)):\n",
    "                        isNewSuperSet = True\n",
    "                    elif (setNew.issubset(setExisting)):\n",
    "                        new_antecedents.remove(list(current_antecedent))\n",
    "                    else:\n",
    "                        pass\n",
    "\n",
    "                \"\"\"\n",
    "                if (x[0] == 'Female'): \n",
    "                    print(\"POST  \", str(list(b)), \"\\t\", rules_result[str(list(b))], \"\\t\", new_antecedents)\n",
    "                \"\"\"\n",
    "                \n",
    "                if(not isNewSuperSet):\n",
    "                    new_antecedents.append(list(a))\n",
    "                rules_result[str(list(b))] = new_antecedents\n",
    " \n",
    "            else:\n",
    "                rules_result[str(list(b))] = [list(a)]\n",
    "            \n",
    "    for consequent, antecedents in rules_result.items():\n",
    "        for antecedent in antecedents:\n",
    "            print(\"Rule: \" + str(antecedent) + \" -> \" + str(consequent))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rule: [] -> ['Adult']\n",
      "Rule: [] -> ['Male']\n",
      "Rule: [] -> ['No']\n",
      "Rule: ['1st'] -> ['Yes']\n",
      "Rule: ['Child'] -> ['Yes']\n",
      "Rule: ['Female'] -> ['Yes']\n",
      "Rule: ['Child'] -> ['3rd']\n",
      "Rule: ['Female', 'No'] -> ['3rd']\n",
      "Rule: ['1st', 'Yes'] -> ['Female']\n",
      "Rule: ['Yes', '2nd'] -> ['Female']\n",
      "Rule: ['Yes', '3rd'] -> ['Female']\n",
      "Rule: ['Male', 'Adult'] -> ['Crew']\n",
      "Rule: ['Male', 'Yes'] -> ['Crew']\n"
     ]
    }
   ],
   "source": [
    "prune_redundant_rules(0.01,0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}