Benchmarks.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Steam Search Engine - Evaluation\n",
    "This document aims to evaluate the performances of the search engine through a series of measures.\n",
    "The set of queries considered in the computations is defined in \"queries.json\".\n",
    "All the measures are applied to the different versions of the project:\n",
    "- base version, ranking model: BM25F;\n",
    "- sentiment analysis v1: every game sentiment value is obtained from the average of its reviews sentiment values;\n",
    "- sentiment analysis v2: every game sentiment value is obatined from the inverted neutral weighted average of its reviews sentiment values.\n",
    "\n",
    "Note:\n",
    "\n",
    "> For the benchmarks to work, it is necessary to create previously the indexes (`inedexdir/base, indexdir/sentiment`).\n",
    "\n",
    "To create all the indexes:\n",
    "`python3 -m setup` or download the pre-created indexes from the GitHub repository.\n",
    "\n",
    "## Benchmarks\n",
    "The following code retrieves the queries set and creates the Benchmarks object `b`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "from BenchmarkUtilities.benchmarks import Benchmarks\n",
    "\n",
    "# file containing benchmark queries\n",
    "file_path = os.path.join(\"BenchmarkUtilities\", \"queries.json\")\n",
    "\n",
    "with open(file_path) as f:\n",
    "    queries = json.load(f)\n",
    "\n",
    "b = Benchmarks(queries)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To have the indexes of each query of the benchmark queries set:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Benchmark queries' indexes:\n",
      "0 1 2 3 4 5 6 7 8 9\n"
     ]
    }
   ],
   "source": [
    "indexes = [i for i in range(len(b.queries))]\n",
    "print(\"Benchmark queries' indexes:\")\n",
    "print(' '.join(map(str, indexes)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some measures are executed on a specific query of the benchmark queries set.  \n",
    "Set the variable `examined_q` to one of the benchmark queries' indexes (printed above) to apply those measures to a chosen query."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "User Information Need: Good and realistic parkour games\n"
     ]
    }
   ],
   "source": [
    "try:\n",
    "    examined_q = 7\n",
    "    print(\"User Information Need: \" + b.queries[examined_q][\"UIN\"])\n",
    "except IndexError as e:\n",
    "    print(e)\n",
    "    print(\"index not valid\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To open the indexes and create the GameSearcher objects for each version:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "All model checkpoint layers were used when initializing TFRobertaForSequenceClassification.\n",
      "\n",
      "All the layers of TFRobertaForSequenceClassification were initialized from the model checkpoint at j-hartmann/emotion-english-distilroberta-base.\n",
      "If your task is similar to the task the model of the checkpoint was trained on, you can already use TFRobertaForSequenceClassification for predictions without further training.\n",
      "All model checkpoint layers were used when initializing TFRobertaForSequenceClassification.\n",
      "\n",
      "All the layers of TFRobertaForSequenceClassification were initialized from the model checkpoint at j-hartmann/emotion-english-distilroberta-base.\n",
      "If your task is similar to the task the model of the checkpoint was trained on, you can already use TFRobertaForSequenceClassification for predictions without further training.\n"
     ]
    }
   ],
   "source": [
    "from whoosh.index import open_dir\n",
    "from MainImplementation.GameSearcher import GameSearcher\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "\n",
    "indexdirBase = \"indexdir/base\"\n",
    "indexdirSentiment = \"indexdir/sentiment\"\n",
    "if (not os.path.exists(indexdirBase) and not os.path.exists(indexdirSentiment)):\n",
    "        raise Exception(\"Index not found\")\n",
    "# create the 2 index objects corresponding to the main index and the sentiment index\n",
    "main_idxb, reviews_idxb = open_dir(indexdirBase + \"/main_index\"), open_dir(indexdirBase + \"/reviews_index\")\n",
    "main_idxs, reviews_idxs = open_dir(indexdirSentiment + \"/main_index\"), open_dir(indexdirSentiment + \"/reviews_index\")\n",
    "\n",
    "# there is one GameSearcher object for every version of the engine\n",
    "SearcherBaseVersion = GameSearcher(main_idxb, reviews_idxb)\n",
    "SearcherSentimentAV = GameSearcher(main_idxs, reviews_idxs, True, \"av\")\n",
    "SearcherSentimentINAV = GameSearcher(main_idxs, reviews_idxs, True, \"inav\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Interpolated Precision\n",
    "Precision of the examined query `examined_q` at every standard recall level."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "UIN: Good and realistic parkour games\n",
      "Query: parkour \\sentiment[very good],\n",
      "Fields: cgt, description\n",
      "Relevant documents: 278360, 17410, 361630, 262390, 300340, 110400, 239140\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# search using the benchmark query and its fields\n",
    "answerB = SearcherBaseVersion.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "answerAV = SearcherSentimentAV.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "answerINAV = SearcherSentimentINAV.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "\n",
    "# exctract the app_id field from the retrieved games\n",
    "aB = [int(x[\"app_id\"]) for x in answerB]\n",
    "aAV = [int(x[\"app_id\"]) for x in answerAV]\n",
    "aINAV = [int(x[\"app_id\"]) for x in answerINAV]\n",
    "\n",
    "# compute the precision at Standard Recall Levels\n",
    "pB = b.precisionSRL(b.precisionNRL(b.queries[examined_q][\"relevant_documents\"], aB))\n",
    "pAV = b.precisionSRL(b.precisionNRL(b.queries[examined_q][\"relevant_documents\"], aAV))\n",
    "pINAV = b.precisionSRL(b.precisionNRL(b.queries[examined_q][\"relevant_documents\"], aINAV))\n",
    "\n",
    "# plot the precisions\n",
    "# apply the default theme\n",
    "sns.set_theme()\n",
    "\n",
    "axes = [\"standard recall levels\", \"precision\"]\n",
    "\n",
    "dfB = pd.DataFrame(pB, columns = axes)\n",
    "dfB[\"Version\"] = \"Base\"\n",
    "\n",
    "dfAV = pd.DataFrame(pAV, columns = axes)\n",
    "dfAV[\"Version\"] = \"Sentiment Analysis AV\"\n",
    "\n",
    "dfINAV = pd.DataFrame(pINAV, columns = axes)\n",
    "dfINAV[\"Version\"] = \"Sentiment Analysis INAV\"\n",
    "\n",
    "# create a dataframe for Seaborn\n",
    "df = pd.concat([dfB, dfAV, dfINAV])\n",
    "\n",
    "# print the examined query\n",
    "b.print_query(examined_q)\n",
    "\n",
    "# plot the line graph\n",
    "pltP = sns.lineplot(data = df, x = 'standard recall levels', y = 'precision', marker='o', markersize=8, \n",
    "             hue=\"Version\", palette=\"colorblind\")\n",
    "\n",
    "# set fixed axes, the semicolon suppress the output\n",
    "pltP.set_xlim([-0.05, 1.05]);\n",
    "pltP.set_ylim([-0.05, 1.05]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Average Precision\n",
    "Average precision at each standard recall level across all queries of the benchmark queries set.  \n",
    "Evaluates overall system performance on a query corpus."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# compute answers list for all the benchmark queries \n",
    "answersB = []\n",
    "answersAV = []\n",
    "answersINAV = []\n",
    "for q in range(len(b.queries)):\n",
    "    answersB.append(SearcherBaseVersion.search(b.queries[q][\"query\"], b.queries[q][\"fields\"]))\n",
    "    answersAV.append(SearcherSentimentAV.search(b.queries[q][\"query\"], b.queries[q][\"fields\"]))\n",
    "    answersINAV.append(SearcherSentimentINAV.search(b.queries[q][\"query\"], b.queries[q][\"fields\"]))\n",
    "\n",
    "# exctract the app_id field from the retrieved games\n",
    "asB = []\n",
    "asAV = []\n",
    "asINAV = []\n",
    "for q in range(len(b.queries)):\n",
    "    asB.append([int(x[\"app_id\"]) for x in answersB[q]])\n",
    "    asAV.append([int(x[\"app_id\"]) for x in answersAV[q]])\n",
    "    asINAV.append([int(x[\"app_id\"]) for x in answersINAV[q]])\n",
    "    \n",
    "# compute the precision at Standard Recall Levels\n",
    "psB = []\n",
    "psAV = []\n",
    "psINAV = []\n",
    "for q in range(len(b.queries)):\n",
    "    psB.append(b.precisionSRL(b.precisionNRL(b.queries[q][\"relevant_documents\"], asB[q])))\n",
    "    psAV.append(b.precisionSRL(b.precisionNRL(b.queries[q][\"relevant_documents\"], asAV[q])))\n",
    "    psINAV.append(b.precisionSRL(b.precisionNRL(b.queries[q][\"relevant_documents\"], asINAV[q])))\n",
    "    \n",
    "# compute the average precision (for each version) over the benchmark queries set\n",
    "avpB = b.AveragePrecision(psB)\n",
    "avpAV = b.AveragePrecision(psAV)\n",
    "avpINAV = b.AveragePrecision(psINAV)\n",
    "\n",
    "# plot the average precisions\n",
    "# apply the default theme\n",
    "sns.set_theme()\n",
    "\n",
    "axes = [\"standard recall levels\", \"average precision\"]\n",
    "\n",
    "dfB = pd.DataFrame(avpB, columns = axes)\n",
    "dfB[\"Version\"] = \"Base\"\n",
    "\n",
    "dfAV = pd.DataFrame(avpAV, columns = axes)\n",
    "dfAV[\"Version\"] = \"Sentiment Analysis AV\"\n",
    "\n",
    "dfINAV = pd.DataFrame(avpINAV, columns = axes)\n",
    "dfINAV[\"Version\"] = \"Sentiment Analysis INAV\"\n",
    "\n",
    "# create a dataframe for Seaborn\n",
    "df = pd.concat([dfB, dfAV, dfINAV])\n",
    "\n",
    "# plot the line graph\n",
    "pltAvp = sns.lineplot(data = df, x = 'standard recall levels', y = 'average precision', marker='o', markersize=8, \n",
    "             hue=\"Version\", palette=\"colorblind\")\n",
    "\n",
    "# set fixed axes, the semicolon suppress the output\n",
    "pltAvp.set_xlim([-0.05, 1.05]); # set x-axis\n",
    "pltAvp.set_ylim([-0.05, 1.05]); # set y-axis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Interpolated Average Precision for an examined query\n",
    "Single average precision value computed on the examined query `examined_q`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "UIN: Good and realistic parkour games\n",
      "Query: parkour \\sentiment[very good],\n",
      "Fields: cgt, description\n",
      "Relevant documents: 278360, 17410, 361630, 262390, 300340, 110400, 239140\n",
      "\n",
      "Average precision: \n",
      "\tBase: 0.2727272727272727\n",
      "\tSentiment Analysis AV: 0.4912587412587413\n",
      "\tSentiment Analysis INAV: 0.6821678321678323\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# search using the benchmark query and its fields\n",
    "answerB = SearcherBaseVersion.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "answerAV = SearcherSentimentAV.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "answerINAV = SearcherSentimentINAV.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "\n",
    "# exctract the app_id field from the retrieved games\n",
    "aB = [int(x[\"app_id\"]) for x in answerB]\n",
    "aAV = [int(x[\"app_id\"]) for x in answerAV]\n",
    "aINAV = [int(x[\"app_id\"]) for x in answerINAV]\n",
    "\n",
    "# compute the interpolated average precision at Standard Recall Levels\n",
    "BIntAvP = b.InterAvgPrec(b.precisionSRL(b.precisionNRL(b.queries[examined_q][\"relevant_documents\"], aB)))\n",
    "AVIntAvP = b.InterAvgPrec(b.precisionSRL(b.precisionNRL(b.queries[examined_q][\"relevant_documents\"], aAV)))\n",
    "INAVIntAvP = b.InterAvgPrec(b.precisionSRL(b.precisionNRL(b.queries[examined_q][\"relevant_documents\"], aINAV)))\n",
    "\n",
    "# print the average precision for each version\n",
    "b.print_query(examined_q)\n",
    "print()\n",
    "print(\"Average precision: \")\n",
    "print(f\"\\tBase: {BIntAvP}\")\n",
    "print(f\"\\tSentiment Analysis AV: {AVIntAvP}\")\n",
    "print(f\"\\tSentiment Analysis INAV: {INAVIntAvP}\")\n",
    "\n",
    "# plot the average precisions\n",
    "# apply the default theme\n",
    "sns.set_theme()\n",
    "\n",
    "versionHUE = [\"Base\", \"Sentiment Analysis AV\", \"Sentiment Analysis INAV\"]\n",
    "versions = [1, 2, 3]\n",
    "AvPs = [BIntAvP, AVIntAvP, INAVIntAvP]\n",
    "\n",
    "# create a dataframe for Seaborn\n",
    "df = pd.DataFrame({\"search-engine version\": versions, \"average precision\": AvPs, \"Version\": versionHUE})\n",
    "\n",
    "# plot the bar graph\n",
    "pltAvps = sns.barplot(data = df, x = \"search-engine version\", y = 'average precision', palette=\"colorblind\", hue=\"Version\")\n",
    "\n",
    "# set fixed axes, the semicolon suppress the output\n",
    "pltAvps.set_ylim([-0.05, 1.05]); # set y-axis\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mean Average Precision (MAP)\n",
    "The average of the average precisions across multiple queries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Average Precision: \n",
      "\tBase: 0.44802750280023007\n",
      "\tSentiment Analysis AV: 0.5997313797313797\n",
      "\tSentiment Analysis INAV: 0.6131237716531834\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# compute answers list for all the benchmark queries \n",
    "answersB = []\n",
    "answersAV = []\n",
    "answersINAV = []\n",
    "for q in range(len(b.queries)):\n",
    "    answersB.append(SearcherBaseVersion.search(b.queries[q][\"query\"], b.queries[q][\"fields\"]))\n",
    "    answersAV.append(SearcherSentimentAV.search(b.queries[q][\"query\"], b.queries[q][\"fields\"]))\n",
    "    answersINAV.append(SearcherSentimentINAV.search(b.queries[q][\"query\"], b.queries[q][\"fields\"]))\n",
    "\n",
    "# exctract the app_id field from the retrieved games\n",
    "asB = []\n",
    "asAV = []\n",
    "asINAV = []\n",
    "for q in range(len(b.queries)):\n",
    "    asB.append([int(x[\"app_id\"]) for x in answersB[q]])\n",
    "    asAV.append([int(x[\"app_id\"]) for x in answersAV[q]])\n",
    "    asINAV.append([int(x[\"app_id\"]) for x in answersINAV[q]])\n",
    "\n",
    "# compute the interpolated average precision at Standard Recall Levels\n",
    "IntAvPsB = []\n",
    "IntAvPsAV = []\n",
    "IntAvPsINAV = []\n",
    "for q in range(len(b.queries)):\n",
    "    IntAvPsB.append(b.InterAvgPrec(b.precisionSRL(b.precisionNRL(b.queries[q][\"relevant_documents\"], asB[q]))))\n",
    "    IntAvPsAV.append(b.InterAvgPrec(b.precisionSRL(b.precisionNRL(b.queries[q][\"relevant_documents\"], asAV[q]))))\n",
    "    IntAvPsINAV.append(b.InterAvgPrec(b.precisionSRL(b.precisionNRL(b.queries[q][\"relevant_documents\"], asINAV[q]))))\n",
    "\n",
    "# compute the Mean Average Precision (for each version) over the benchmark queries set\n",
    "MapB = b.MAP(IntAvPsB)\n",
    "MapAV = b.MAP(IntAvPsAV)\n",
    "MapINAV = b.MAP(IntAvPsINAV)\n",
    "\n",
    "# print the MAP for each version\n",
    "print(\"Mean Average Precision: \")\n",
    "print(f\"\\tBase: {MapB}\")\n",
    "print(f\"\\tSentiment Analysis AV: {MapAV}\")\n",
    "print(f\"\\tSentiment Analysis INAV: {MapINAV}\")\n",
    "\n",
    "# plot mean average precisions\n",
    "# apply the default theme\n",
    "sns.set_theme()\n",
    "\n",
    "versionHUE = [\"Base\", \"Sentiment Analysis AV\", \"Sentiment Analysis INAV\"]\n",
    "versions = [1, 2, 3]\n",
    "Maps = [MapB, MapAV, MapINAV]\n",
    "\n",
    "# create a dataframe for Seaborn\n",
    "df = pd.DataFrame({\"search-engine version\": versions, \"mean average precision\": Maps, \"Version\": versionHUE})\n",
    "\n",
    "# plot the bar graph\n",
    "pltMap = sns.barplot(data = df, x = \"search-engine version\", y = 'mean average precision', palette=\"colorblind\", hue=\"Version\")\n",
    "\n",
    "# set fixed axes, the semicolon suppress the output\n",
    "pltMap.set_ylim([-0.05, 1.05]); # set y-axis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### R-Precision\n",
    "Precision at R-th position in the ranking of results for a query that has R relevant documents."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# compute answers list for all the benchmark queries \n",
    "answersB = []\n",
    "answersAV = []\n",
    "answersINAV = []\n",
    "for q in range(len(b.queries)):\n",
    "    answersB.append(SearcherBaseVersion.search(b.queries[q][\"query\"], b.queries[q][\"fields\"]))\n",
    "    answersAV.append(SearcherSentimentAV.search(b.queries[q][\"query\"], b.queries[q][\"fields\"]))\n",
    "    answersINAV.append(SearcherSentimentINAV.search(b.queries[q][\"query\"], b.queries[q][\"fields\"]))\n",
    "    \n",
    "# exctract the app_id field from the retrieved games\n",
    "asB = []\n",
    "asAV = []\n",
    "asINAV = []\n",
    "for q in range(len(b.queries)):\n",
    "    asB.append([int(x[\"app_id\"]) for x in answersB[q]])\n",
    "    asAV.append([int(x[\"app_id\"]) for x in answersAV[q]])\n",
    "    asINAV.append([int(x[\"app_id\"]) for x in answersINAV[q]])\n",
    "    \n",
    "# create a list with all the versions' answers\n",
    "AsVersions = {\n",
    "    \"Base\" : asB,\n",
    "    \"AV\" : asAV,\n",
    "    \"INAV\" : asINAV\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Set `v1` and `v2` variables to the versions of the search engine whose R-precisions you want to compare."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "v1 = \"Base\"\n",
    "v2 = \"INAV\"\n",
    "\n",
    "if (v1 not in AsVersions or v2 not in AsVersions):\n",
    "    raise Exception(\"Version not valid\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compute and plot R-precision histogram.  \n",
    "For the queries that do not have sentiment part, the R-precisions' subtraction will be zero, since the output is the same for Base version and Sentiment versions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "RPs = b.RPrecisions(AsVersions[v1], AsVersions[v2])\n",
    "\n",
    "# plot the R precisions\n",
    "# apply the default theme\n",
    "sns.set_theme()\n",
    "\n",
    "# create a DataFrame for Seaborn\n",
    "df = pd.DataFrame({\"query number\": [x for x in range(len(b.queries))], \"R-Precision v1/v2\": RPs})\n",
    "\n",
    "# use Seaborn to create a bar plot\n",
    "pltRP = sns.barplot(data=df, x='query number', y='R-Precision v1/v2', color=\"skyblue\")\n",
    "\n",
    "# add a line to separate the positive area to the negative one\n",
    "pltRP.axhline(0, color='k', clip_on=False)\n",
    "\n",
    "# set fixed axes, the semicolon suppress the output\n",
    "pltRP.set_ylim([-1.05, 1.05]);  # Set y-axis limits"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Harmonic Mean (F-Measure)\n",
    "Single value measure which combines recall and precision of the examined query `examined_q`.  \n",
    "High value only when both recall and precision are high."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "UIN: Good and realistic parkour games\n",
      "Query: parkour \\sentiment[very good],\n",
      "Fields: cgt, description\n",
      "Relevant documents: 278360, 17410, 361630, 262390, 300340, 110400, 239140\n",
      "\n",
      "Harmonic mean: \n",
      "\tBase: 0.14814814814814817\n",
      "\tSentiment Analysis AV: 0.4444444444444444\n",
      "\tSentiment Analysis INAV: 0.4444444444444444\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# search using the benchmark query and its fields\n",
    "answerB = SearcherBaseVersion.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "answerAV = SearcherSentimentAV.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "answerINAV = SearcherSentimentINAV.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "\n",
    "# exctract the app_id field from the retrieved games\n",
    "aB = [int(x[\"app_id\"]) for x in answerB]\n",
    "aAV = [int(x[\"app_id\"]) for x in answerAV]\n",
    "aINAV = [int(x[\"app_id\"]) for x in answerINAV]\n",
    "\n",
    "# compute the F-Measure (Harmonic Mean)\n",
    "FmB = b.FMeasure(b.queries[examined_q][\"relevant_documents\"], aB)\n",
    "FmAV = b.FMeasure(b.queries[examined_q][\"relevant_documents\"], aAV)\n",
    "FmINAV = b.FMeasure(b.queries[examined_q][\"relevant_documents\"], aINAV)\n",
    "\n",
    "# print the average precision for each version\n",
    "b.print_query(examined_q)\n",
    "print()\n",
    "print(\"Harmonic mean: \")\n",
    "print(f\"\\tBase: {FmB}\")\n",
    "print(f\"\\tSentiment Analysis AV: {FmAV}\")\n",
    "print(f\"\\tSentiment Analysis INAV: {FmINAV}\")\n",
    "\n",
    "# plot the average precisions\n",
    "# apply the default theme\n",
    "sns.set_theme()\n",
    "\n",
    "versionHUE = [\"Base\", \"Sentiment Analysis AV\", \"Sentiment Analysis INAV\"]\n",
    "versions = [1, 2, 3]\n",
    "Fms = [FmB, FmAV, FmINAV]\n",
    "\n",
    "\n",
    "# create a dataframe for Seaborn\n",
    "df = pd.DataFrame({\"search-engine version\": versions, \"harmonic mean\": Fms, \"Version\": versionHUE})\n",
    "\n",
    "# plot the bar graph\n",
    "pltFm = sns.barplot(data = df, x = \"search-engine version\", y = 'harmonic mean', palette=\"colorblind\", hue=\"Version\")\n",
    "\n",
    "# set fixed axes, the semicolon suppress the output\n",
    "pltFm.set_ylim([-0.05, 1.05]); # set y-axis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### E-Measure\n",
    "It allows the user to specify whether they are more interested in recall or in precision.  \n",
    "Modify the `EMeasure_b` variable to change wheights of recall and precision: \n",
    "- `EMeasure_b = 1` to get the complement of harmonic mean\n",
    "- `EMeasure_b > 1` to emphasize precision\n",
    "- `0 <= EMeasure_b < 1` to emphasize recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "UIN: Good and realistic parkour games\n",
      "Query: parkour \\sentiment[very good],\n",
      "Fields: cgt, description\n",
      "Relevant documents: 278360, 17410, 361630, 262390, 300340, 110400, 239140\n",
      "\n",
      "E-Measure: \n",
      "\tBase: 0.8850574712643678\n",
      "\tSentiment Analysis AV: 0.6551724137931034\n",
      "\tSentiment Analysis INAV: 0.6551724137931034\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "EMeasure_b = 0.5\n",
    "\n",
    "# search using the benchmark query and its fields\n",
    "answerB = SearcherBaseVersion.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "answerAV = SearcherSentimentAV.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "answerINAV = SearcherSentimentINAV.search(b.queries[examined_q][\"query\"], b.queries[examined_q][\"fields\"])\n",
    "\n",
    "# exctract the app_id field from the retrieved games\n",
    "aB = [int(x[\"app_id\"]) for x in answerB]\n",
    "aAV = [int(x[\"app_id\"]) for x in answerAV]\n",
    "aINAV = [int(x[\"app_id\"]) for x in answerINAV]\n",
    "\n",
    "# compute the E-Measure\n",
    "EmB = b.EMeasure(b.queries[examined_q][\"relevant_documents\"], aB, EMeasure_b)\n",
    "EmAV = b.EMeasure(b.queries[examined_q][\"relevant_documents\"], aAV, EMeasure_b)\n",
    "EmINAV = b.EMeasure(b.queries[examined_q][\"relevant_documents\"], aINAV, EMeasure_b)\n",
    "\n",
    "# print the average precision for each version\n",
    "b.print_query(examined_q)\n",
    "print()\n",
    "print(\"E-Measure: \")\n",
    "print(f\"\\tBase: {EmB}\")\n",
    "print(f\"\\tSentiment Analysis AV: {EmAV}\")\n",
    "print(f\"\\tSentiment Analysis INAV: {EmINAV}\")\n",
    "\n",
    "# plot the average precisions\n",
    "# apply the default theme\n",
    "sns.set_theme()\n",
    "\n",
    "versionHUE = [\"Base\", \"Sentiment Analysis AV\", \"Sentiment Analysis INAV\"]\n",
    "versions = [1, 2, 3]\n",
    "Ems = [EmB, EmAV, EmINAV]\n",
    "\n",
    "# create a dataframe for Seaborn\n",
    "df = pd.DataFrame({\"search-engine version\": versions, \"e-measure\": Ems, \"Version\": versionHUE})\n",
    "\n",
    "# plot the bar graph\n",
    "pltEm = sns.barplot(data = df, x = \"search-engine version\", y = 'e-measure', palette=\"colorblind\", hue=\"Version\")\n",
    "\n",
    "# set fixed axes, the semicolon suppress the output\n",
    "pltEm.set_ylim([-0.05, 1.05]); # set y-axis"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}