Experiment 1 Toy Example.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "ced436c8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.utils import shuffle\n",
    "from rules import standard_deviation_rankings, divisiveness, win_rate, bootstrap3, bootstrap4, copeland, divisiveness_copeland, using_preflib, data_transform\n",
    "import math\n",
    "import warnings; warnings.simplefilter('ignore')\n",
    "def get_kendall(x, c1, c2):\n",
    "    return pd.DataFrame([x[c1],x[c2]]).T.corr(method='kendall')[1].values[0]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b81b6602",
   "metadata": {},
   "source": [
    "# This is a toy example for the experiment 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "825b20e3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialize your experiments\n",
    "METHOD = 'UM10'\n",
    "N_PROFILES = 1\n",
    "N_MINIMUM_ISSUES = 3\n",
    "N_MAXIMUM_ISSUES = 18\n",
    "RANGE_NUMBER_AGENTS = [100]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "ba9f98d1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running profile:  0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 6/6 [00:00<00:00, 194.87it/s]\n",
      "100%|██████████| 6/6 [00:00<00:00, 186.94it/s]\n",
      "100%|██████████| 12/12 [00:00<00:00, 188.56it/s]\n",
      "100%|██████████| 12/12 [00:00<00:00, 194.03it/s]\n",
      "100%|██████████| 20/20 [00:00<00:00, 196.93it/s]\n",
      "100%|██████████| 20/20 [00:00<00:00, 208.25it/s]\n",
      "100%|██████████| 30/30 [00:00<00:00, 195.93it/s]\n",
      "100%|██████████| 30/30 [00:00<00:00, 211.01it/s]\n",
      "100%|██████████| 42/42 [00:00<00:00, 192.30it/s]\n",
      "100%|██████████| 42/42 [00:00<00:00, 189.51it/s]\n",
      "100%|██████████| 56/56 [00:00<00:00, 185.36it/s]\n",
      "100%|██████████| 56/56 [00:00<00:00, 192.65it/s]\n",
      "100%|██████████| 72/72 [00:00<00:00, 177.33it/s]\n",
      "100%|██████████| 72/72 [00:00<00:00, 183.95it/s]\n",
      "100%|██████████| 90/90 [00:00<00:00, 171.41it/s]\n",
      "100%|██████████| 90/90 [00:00<00:00, 179.08it/s]\n",
      "100%|██████████| 110/110 [00:00<00:00, 164.46it/s]\n",
      "100%|██████████| 110/110 [00:00<00:00, 164.59it/s]\n",
      "100%|██████████| 132/132 [00:00<00:00, 152.95it/s]\n",
      "100%|██████████| 132/132 [00:00<00:00, 158.10it/s]\n",
      "100%|██████████| 156/156 [00:01<00:00, 144.81it/s]\n",
      "100%|██████████| 156/156 [00:01<00:00, 144.29it/s]\n",
      "100%|██████████| 182/182 [00:01<00:00, 137.21it/s]\n",
      "100%|██████████| 182/182 [00:01<00:00, 142.35it/s]\n",
      "100%|██████████| 210/210 [00:01<00:00, 133.88it/s]\n",
      "100%|██████████| 210/210 [00:01<00:00, 134.47it/s]\n",
      "100%|██████████| 240/240 [00:01<00:00, 128.09it/s]\n",
      "100%|██████████| 240/240 [00:01<00:00, 127.80it/s]\n",
      "100%|██████████| 272/272 [00:02<00:00, 121.12it/s]\n",
      "100%|██████████| 272/272 [00:02<00:00, 125.01it/s]\n"
     ]
    }
   ],
   "source": [
    "results = []\n",
    "\n",
    "for profile in range(N_PROFILES):\n",
    "    \n",
    "    print(\"Running profile: \", profile)\n",
    "    \n",
    "    for number_issues in range(N_MINIMUM_ISSUES,N_MAXIMUM_ISSUES+1):\n",
    "\n",
    "        for number_agents in RANGE_NUMBER_AGENTS:\n",
    "            \n",
    "            # Create the population using the selected method and the number of issues and agents\n",
    "            all_rankings = using_preflib(number_issues, number_agents, type_=METHOD)\n",
    "            \n",
    "            # Transform the rankings into pairwise comparisons \n",
    "            # and put in the correct format for running the voting rule\n",
    "            output = data_transform(all_rankings)\n",
    "            data = pd.concat(output).reset_index(drop=True)\n",
    "            data[\"id\"] = range(1, data.shape[0] + 1)\n",
    "            data[\"selected\"] = data[\"option_a\"]\n",
    "            data[\"option_a_sorted\"] = data[[\"option_a\", \"option_b\"]].min(axis=1).astype(int)\n",
    "            data[\"option_b_sorted\"] = data[[\"option_a\", \"option_b\"]].max(axis=1).astype(int)\n",
    "            data[\"card_id\"] = data[\"option_a_sorted\"].astype(str) + \"_\" + data[\"option_b_sorted\"].astype(str)\n",
    "            a = data[[\"option_a\", \"option_b\", \"selected\"]].values\n",
    "            data[\"option_source\"] = np.where(a[:, 1] == a[:, 2], a[:, 0], a[:, 1])\n",
    "            data[\"option_target\"] = np.where(a[:, 0] == a[:, 2], a[:, 0], a[:, 1])\n",
    "            \n",
    "            # Run each voting rule with the created profile\n",
    "            df_win_rate = bootstrap4(data, win_rate, iterations=1, \n",
    "                                     aggregate=False, rank=True, rank_column=\"rank\")\n",
    "            ranked_win_rate = list(df_win_rate.sort_values(by='value', ascending=False)['id'].values)\n",
    "    \n",
    "            df_copeland = bootstrap4(data, copeland, iterations=1, \n",
    "                                     aggregate=False, rank=True, rank_column=\"rank\")\n",
    "            ranked_copeland = list(df_copeland.sort_values(by='value', ascending=False)['id'].values)\n",
    "\n",
    "            df_divisiveness_win_rate = bootstrap4(data, divisiveness, iterations=1, \n",
    "                                         aggregate=False, rank=True, rank_column=\"rank\")\n",
    "            ranked_divisiveness_win_rate = list(df_divisiveness_win_rate.sort_values(by='value', ascending=False)['id'].values)\n",
    "    \n",
    "            df_divisiveness_copeland = bootstrap4(data, divisiveness_copeland, iterations=1, \n",
    "                                         aggregate=False, rank=True, rank_column=\"rank\")\n",
    "            ranked_divisiveness_copeland = list(df_divisiveness_copeland.sort_values(by='value', ascending=False)['id'].values)\n",
    "\n",
    "            df_std = standard_deviation_rankings(all_rankings)\n",
    "            ranked_std = list(df_std.sort_values(by='value', ascending=False)['id'].values)\n",
    "\n",
    "            del all_rankings \n",
    "            results.append([\n",
    "                            number_issues, \n",
    "                            number_agents,\n",
    "                            profile,\n",
    "                            ranked_win_rate,\n",
    "                            ranked_copeland,\n",
    "                            ranked_std,\n",
    "                            ranked_divisiveness_win_rate, \n",
    "                            ranked_divisiveness_copeland,])\n",
    "\n",
    "    plot = pd.DataFrame(results, columns=['number_issues',\n",
    "                                          'number_agents',\n",
    "                                          'profile_id',\n",
    "                                          'RANK(WR)',\n",
    "                                          'RANK(COP)',\n",
    "                                          'RANK(STD)',\n",
    "                                          'RANK(DIV(WR))',\n",
    "                                          'RANK(DIV(COP))'])\n",
    "    \n",
    "    plot.to_csv(\"data/results_complete_method_%s_profile_%s.csv\"%(METHOD,N_PROFILES), index=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "e8e9ad14",
   "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>number_issues</th>\n",
       "      <th>number_agents</th>\n",
       "      <th>number_profiles</th>\n",
       "      <th>RANK(WR)</th>\n",
       "      <th>RANK(COP)</th>\n",
       "      <th>RANK(STD)</th>\n",
       "      <th>RANK(DIV(WR))</th>\n",
       "      <th>RANK(DIV(COP))</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[0, 2, 1]</td>\n",
       "      <td>[0, 2, 1]</td>\n",
       "      <td>[1, 0, 2]</td>\n",
       "      <td>[1, 0, 2]</td>\n",
       "      <td>[1, 0, 2]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[1, 2, 3, 0]</td>\n",
       "      <td>[1, 2, 3, 0]</td>\n",
       "      <td>[1, 2, 0, 3]</td>\n",
       "      <td>[1, 2, 0, 3]</td>\n",
       "      <td>[1, 2, 0, 3]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[3, 0, 4, 2, 1]</td>\n",
       "      <td>[3, 4, 0, 2, 1]</td>\n",
       "      <td>[0, 4, 1, 2, 3]</td>\n",
       "      <td>[0, 3, 4, 1, 2]</td>\n",
       "      <td>[0, 1, 4, 2, 3]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[1, 2, 5, 3, 0, 4]</td>\n",
       "      <td>[1, 5, 2, 0, 3, 4]</td>\n",
       "      <td>[0, 5, 1, 3, 4, 2]</td>\n",
       "      <td>[0, 5, 4, 1, 3, 2]</td>\n",
       "      <td>[0, 5, 1, 4, 3, 2]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[2, 6, 3, 0, 5, 4, 1]</td>\n",
       "      <td>[2, 6, 5, 0, 1, 3, 4]</td>\n",
       "      <td>[2, 5, 3, 4, 6, 1, 0]</td>\n",
       "      <td>[2, 5, 3, 4, 1, 6, 0]</td>\n",
       "      <td>[2, 5, 4, 3, 6, 1, 0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>8</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[4, 6, 7, 5, 0, 1, 3, 2]</td>\n",
       "      <td>[6, 4, 7, 1, 3, 5, 0, 2]</td>\n",
       "      <td>[0, 3, 1, 7, 2, 5, 6, 4]</td>\n",
       "      <td>[0, 3, 1, 7, 2, 5, 6, 4]</td>\n",
       "      <td>[0, 1, 3, 2, 7, 6, 5, 4]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>9</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[5, 8, 6, 1, 0, 3, 2, 7, 4]</td>\n",
       "      <td>[5, 6, 8, 0, 3, 1, 7, 2, 4]</td>\n",
       "      <td>[5, 8, 6, 0, 1, 3, 2, 7, 4]</td>\n",
       "      <td>[5, 8, 6, 0, 1, 3, 2, 4, 7]</td>\n",
       "      <td>[5, 8, 6, 1, 0, 3, 7, 4, 2]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[9, 1, 8, 7, 6, 4, 2, 3, 5, 0]</td>\n",
       "      <td>[1, 9, 8, 6, 7, 4, 5, 2, 3, 0]</td>\n",
       "      <td>[4, 8, 7, 9, 0, 3, 2, 1, 5, 6]</td>\n",
       "      <td>[8, 4, 9, 7, 0, 3, 2, 1, 5, 6]</td>\n",
       "      <td>[7, 8, 9, 4, 0, 1, 3, 2, 5, 6]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>11</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[5, 0, 8, 6, 3, 4, 9, 7, 10, 2, 1]</td>\n",
       "      <td>[0, 5, 6, 3, 4, 8, 9, 7, 10, 2, 1]</td>\n",
       "      <td>[0, 1, 10, 4, 2, 8, 6, 5, 3, 7, 9]</td>\n",
       "      <td>[0, 1, 10, 2, 4, 5, 8, 6, 3, 7, 9]</td>\n",
       "      <td>[0, 1, 10, 5, 6, 2, 4, 3, 8, 7, 9]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>12</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[4, 11, 9, 1, 0, 8, 6, 5, 2, 10, 7, 3]</td>\n",
       "      <td>[4, 9, 11, 1, 8, 0, 6, 5, 2, 3, 10, 7]</td>\n",
       "      <td>[7, 6, 2, 9, 11, 10, 5, 1, 0, 3, 4, 8]</td>\n",
       "      <td>[7, 2, 6, 9, 11, 10, 5, 1, 3, 4, 0, 8]</td>\n",
       "      <td>[7, 9, 6, 2, 11, 10, 1, 5, 0, 3, 4, 8]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   number_issues  number_agents  number_profiles  \\\n",
       "0              3            100                0   \n",
       "1              4            100                0   \n",
       "2              5            100                0   \n",
       "3              6            100                0   \n",
       "4              7            100                0   \n",
       "5              8            100                0   \n",
       "6              9            100                0   \n",
       "7             10            100                0   \n",
       "8             11            100                0   \n",
       "9             12            100                0   \n",
       "\n",
       "                                 RANK(WR)  \\\n",
       "0                               [0, 2, 1]   \n",
       "1                            [1, 2, 3, 0]   \n",
       "2                         [3, 0, 4, 2, 1]   \n",
       "3                      [1, 2, 5, 3, 0, 4]   \n",
       "4                   [2, 6, 3, 0, 5, 4, 1]   \n",
       "5                [4, 6, 7, 5, 0, 1, 3, 2]   \n",
       "6             [5, 8, 6, 1, 0, 3, 2, 7, 4]   \n",
       "7          [9, 1, 8, 7, 6, 4, 2, 3, 5, 0]   \n",
       "8      [5, 0, 8, 6, 3, 4, 9, 7, 10, 2, 1]   \n",
       "9  [4, 11, 9, 1, 0, 8, 6, 5, 2, 10, 7, 3]   \n",
       "\n",
       "                                RANK(COP)  \\\n",
       "0                               [0, 2, 1]   \n",
       "1                            [1, 2, 3, 0]   \n",
       "2                         [3, 4, 0, 2, 1]   \n",
       "3                      [1, 5, 2, 0, 3, 4]   \n",
       "4                   [2, 6, 5, 0, 1, 3, 4]   \n",
       "5                [6, 4, 7, 1, 3, 5, 0, 2]   \n",
       "6             [5, 6, 8, 0, 3, 1, 7, 2, 4]   \n",
       "7          [1, 9, 8, 6, 7, 4, 5, 2, 3, 0]   \n",
       "8      [0, 5, 6, 3, 4, 8, 9, 7, 10, 2, 1]   \n",
       "9  [4, 9, 11, 1, 8, 0, 6, 5, 2, 3, 10, 7]   \n",
       "\n",
       "                                RANK(STD)  \\\n",
       "0                               [1, 0, 2]   \n",
       "1                            [1, 2, 0, 3]   \n",
       "2                         [0, 4, 1, 2, 3]   \n",
       "3                      [0, 5, 1, 3, 4, 2]   \n",
       "4                   [2, 5, 3, 4, 6, 1, 0]   \n",
       "5                [0, 3, 1, 7, 2, 5, 6, 4]   \n",
       "6             [5, 8, 6, 0, 1, 3, 2, 7, 4]   \n",
       "7          [4, 8, 7, 9, 0, 3, 2, 1, 5, 6]   \n",
       "8      [0, 1, 10, 4, 2, 8, 6, 5, 3, 7, 9]   \n",
       "9  [7, 6, 2, 9, 11, 10, 5, 1, 0, 3, 4, 8]   \n",
       "\n",
       "                            RANK(DIV(WR))  \\\n",
       "0                               [1, 0, 2]   \n",
       "1                            [1, 2, 0, 3]   \n",
       "2                         [0, 3, 4, 1, 2]   \n",
       "3                      [0, 5, 4, 1, 3, 2]   \n",
       "4                   [2, 5, 3, 4, 1, 6, 0]   \n",
       "5                [0, 3, 1, 7, 2, 5, 6, 4]   \n",
       "6             [5, 8, 6, 0, 1, 3, 2, 4, 7]   \n",
       "7          [8, 4, 9, 7, 0, 3, 2, 1, 5, 6]   \n",
       "8      [0, 1, 10, 2, 4, 5, 8, 6, 3, 7, 9]   \n",
       "9  [7, 2, 6, 9, 11, 10, 5, 1, 3, 4, 0, 8]   \n",
       "\n",
       "                           RANK(DIV(COP))  \n",
       "0                               [1, 0, 2]  \n",
       "1                            [1, 2, 0, 3]  \n",
       "2                         [0, 1, 4, 2, 3]  \n",
       "3                      [0, 5, 1, 4, 3, 2]  \n",
       "4                   [2, 5, 4, 3, 6, 1, 0]  \n",
       "5                [0, 1, 3, 2, 7, 6, 5, 4]  \n",
       "6             [5, 8, 6, 1, 0, 3, 7, 4, 2]  \n",
       "7          [7, 8, 9, 4, 0, 1, 3, 2, 5, 6]  \n",
       "8      [0, 1, 10, 5, 6, 2, 4, 3, 8, 7, 9]  \n",
       "9  [7, 9, 6, 2, 11, 10, 1, 5, 0, 3, 4, 8]  "
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "1ccb9f43",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"data/results_complete_method_%s_profile_%s.csv\"%(METHOD,N_PROFILES))\n",
    "df['RANK(WR)'] = df['RANK(WR)'].apply(lambda x: eval(x))\n",
    "df['RANK(STD)'] = df['RANK(STD)'].apply(lambda x: eval(x))\n",
    "df['RANK(COP)'] = df['RANK(COP)'].apply(lambda x: eval(x))\n",
    "df['RANK(DIV(WR))'] = df['RANK(DIV(WR))'].apply(lambda x: eval(x))\n",
    "df['RANK(DIV(COP))'] = df['RANK(DIV(COP))'].apply(lambda x: eval(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "63a97b42",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['KT(WR,STD)'] = df.apply(lambda x: get_kendall(x, c1='RANK(WR)', c2='RANK(STD)'), axis=1)\n",
    "df['KT(WR,COP)'] = df.apply(lambda x: get_kendall(x, c1='RANK(WR)', c2='RANK(COP)'), axis=1)\n",
    "df['KT(WR,RANK(DIV(WR)))'] = df.apply(lambda x: get_kendall(x, c1='RANK(WR)', c2='RANK(DIV(WR))'), axis=1)\n",
    "df['KT(WR,RANK(DIV(COP)))'] = df.apply(lambda x: get_kendall(x, c1='RANK(WR)', c2='RANK(DIV(COP))'), axis=1)\n",
    "df['KT(RANK(DIV(WR)),STD)'] = df.apply(lambda x: get_kendall(x, c1='RANK(DIV(WR))', c2='RANK(STD)'), axis=1)\n",
    "df['KT(RANK(DIV(WR)),COP)'] = df.apply(lambda x: get_kendall(x, c1='RANK(DIV(WR))', c2='RANK(COP)'), axis=1)\n",
    "df['KT(RANK(DIV(WR)),RANK(DIV(COP)))'] = df.apply(lambda x: get_kendall(x, c1='RANK(DIV(WR))', c2='RANK(DIV(COP))'), axis=1)\n",
    "df['KT(RANK(DIV(COP)),STD)'] = df.apply(lambda x: get_kendall(x, c1='RANK(DIV(COP))', c2='RANK(STD)'), axis=1)\n",
    "df['KT(RANK(DIV(COP)),COP)'] = df.apply(lambda x: get_kendall(x, c1='RANK(DIV(COP))', c2='RANK(COP)'), axis=1)\n",
    "df['KT(RANK(DIV(COP)),RANK(DIV(WR)))'] = df.apply(lambda x: get_kendall(x, c1='RANK(DIV(COP))', c2='RANK(DIV(WR))'), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "815319bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv(\"data/results_complete_method_%s_profile_%s_kendall.csv\"%(METHOD,N_PROFILES), index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "321c5332",
   "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>number_issues</th>\n",
       "      <th>number_agents</th>\n",
       "      <th>number_profiles</th>\n",
       "      <th>RANK(WR)</th>\n",
       "      <th>RANK(COP)</th>\n",
       "      <th>RANK(STD)</th>\n",
       "      <th>RANK(DIV(WR))</th>\n",
       "      <th>RANK(DIV(COP))</th>\n",
       "      <th>KT(WR,STD)</th>\n",
       "      <th>KT(WR,COP)</th>\n",
       "      <th>KT(WR,RANK(DIV(WR)))</th>\n",
       "      <th>KT(WR,RANK(DIV(COP)))</th>\n",
       "      <th>KT(RANK(DIV(WR)),STD)</th>\n",
       "      <th>KT(RANK(DIV(WR)),COP)</th>\n",
       "      <th>KT(RANK(DIV(WR)),RANK(DIV(COP)))</th>\n",
       "      <th>KT(RANK(DIV(COP)),STD)</th>\n",
       "      <th>KT(RANK(DIV(COP)),COP)</th>\n",
       "      <th>KT(RANK(DIV(COP)),RANK(DIV(WR)))</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[0, 2, 1]</td>\n",
       "      <td>[0, 2, 1]</td>\n",
       "      <td>[1, 0, 2]</td>\n",
       "      <td>[1, 0, 2]</td>\n",
       "      <td>[1, 0, 2]</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[1, 2, 3, 0]</td>\n",
       "      <td>[1, 2, 3, 0]</td>\n",
       "      <td>[1, 2, 0, 3]</td>\n",
       "      <td>[1, 2, 0, 3]</td>\n",
       "      <td>[1, 2, 0, 3]</td>\n",
       "      <td>-0.666667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.666667</td>\n",
       "      <td>-0.666667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.666667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.666667</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[3, 0, 4, 2, 1]</td>\n",
       "      <td>[3, 4, 0, 2, 1]</td>\n",
       "      <td>[0, 4, 1, 2, 3]</td>\n",
       "      <td>[0, 3, 4, 1, 2]</td>\n",
       "      <td>[0, 1, 4, 2, 3]</td>\n",
       "      <td>-0.800000</td>\n",
       "      <td>-0.400000</td>\n",
       "      <td>-0.200000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>-0.400000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.800000</td>\n",
       "      <td>0.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[1, 2, 5, 3, 0, 4]</td>\n",
       "      <td>[1, 5, 2, 0, 3, 4]</td>\n",
       "      <td>[0, 5, 1, 3, 4, 2]</td>\n",
       "      <td>[0, 5, 4, 1, 3, 2]</td>\n",
       "      <td>[0, 5, 1, 4, 3, 2]</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>-0.066667</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>-0.200000</td>\n",
       "      <td>0.466667</td>\n",
       "      <td>0.466667</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.866667</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>[2, 6, 3, 0, 5, 4, 1]</td>\n",
       "      <td>[2, 6, 5, 0, 1, 3, 4]</td>\n",
       "      <td>[2, 5, 3, 4, 6, 1, 0]</td>\n",
       "      <td>[2, 5, 3, 4, 1, 6, 0]</td>\n",
       "      <td>[2, 5, 4, 3, 6, 1, 0]</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.238095</td>\n",
       "      <td>0.428571</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.047619</td>\n",
       "      <td>0.904762</td>\n",
       "      <td>-0.047619</td>\n",
       "      <td>0.047619</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   number_issues  number_agents  number_profiles               RANK(WR)  \\\n",
       "0              3            100                0              [0, 2, 1]   \n",
       "1              4            100                0           [1, 2, 3, 0]   \n",
       "2              5            100                0        [3, 0, 4, 2, 1]   \n",
       "3              6            100                0     [1, 2, 5, 3, 0, 4]   \n",
       "4              7            100                0  [2, 6, 3, 0, 5, 4, 1]   \n",
       "\n",
       "               RANK(COP)              RANK(STD)          RANK(DIV(WR))  \\\n",
       "0              [0, 2, 1]              [1, 0, 2]              [1, 0, 2]   \n",
       "1           [1, 2, 3, 0]           [1, 2, 0, 3]           [1, 2, 0, 3]   \n",
       "2        [3, 4, 0, 2, 1]        [0, 4, 1, 2, 3]        [0, 3, 4, 1, 2]   \n",
       "3     [1, 5, 2, 0, 3, 4]     [0, 5, 1, 3, 4, 2]     [0, 5, 4, 1, 3, 2]   \n",
       "4  [2, 6, 5, 0, 1, 3, 4]  [2, 5, 3, 4, 6, 1, 0]  [2, 5, 3, 4, 1, 6, 0]   \n",
       "\n",
       "          RANK(DIV(COP))  KT(WR,STD)  KT(WR,COP)  KT(WR,RANK(DIV(WR)))  \\\n",
       "0              [1, 0, 2]   -0.333333    1.000000             -0.333333   \n",
       "1           [1, 2, 0, 3]   -0.666667    1.000000             -0.666667   \n",
       "2        [0, 1, 4, 2, 3]   -0.800000   -0.400000             -0.200000   \n",
       "3     [0, 5, 1, 4, 3, 2]   -0.333333   -0.066667              0.200000   \n",
       "4  [2, 5, 4, 3, 6, 1, 0]    0.333333    0.333333              0.238095   \n",
       "\n",
       "   KT(WR,RANK(DIV(COP)))  KT(RANK(DIV(WR)),STD)  KT(RANK(DIV(WR)),COP)  \\\n",
       "0              -0.333333               1.000000              -0.333333   \n",
       "1              -0.666667               1.000000              -0.666667   \n",
       "2               0.200000               0.400000              -0.400000   \n",
       "3              -0.200000               0.466667               0.466667   \n",
       "4               0.428571               0.142857               0.142857   \n",
       "\n",
       "   KT(RANK(DIV(WR)),RANK(DIV(COP)))  KT(RANK(DIV(COP)),STD)  \\\n",
       "0                          1.000000                1.000000   \n",
       "1                          1.000000                1.000000   \n",
       "2                          0.600000                0.000000   \n",
       "3                          0.333333                0.866667   \n",
       "4                          0.047619                0.904762   \n",
       "\n",
       "   KT(RANK(DIV(COP)),COP)  KT(RANK(DIV(COP)),RANK(DIV(WR)))  \n",
       "0               -0.333333                          1.000000  \n",
       "1               -0.666667                          1.000000  \n",
       "2               -0.800000                          0.600000  \n",
       "3                0.333333                          0.333333  \n",
       "4               -0.047619                          0.047619  "
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "958c29fe",
   "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>number_issues</th>\n",
       "      <th>number_profiles</th>\n",
       "      <th>number_agents</th>\n",
       "      <th>variable</th>\n",
       "      <th>value</th>\n",
       "      <th>Comparison</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>100</td>\n",
       "      <td>KT(RANK(DIV(WR)),RANK(DIV(COP)))</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>100</td>\n",
       "      <td>KT(RANK(DIV(WR)),RANK(DIV(COP)))</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>100</td>\n",
       "      <td>KT(RANK(DIV(WR)),RANK(DIV(COP)))</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>100</td>\n",
       "      <td>KT(RANK(DIV(WR)),RANK(DIV(COP)))</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>100</td>\n",
       "      <td>KT(RANK(DIV(WR)),RANK(DIV(COP)))</td>\n",
       "      <td>0.047619</td>\n",
       "      <td>KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   number_issues  number_profiles  number_agents  \\\n",
       "0              3                0            100   \n",
       "1              4                0            100   \n",
       "2              5                0            100   \n",
       "3              6                0            100   \n",
       "4              7                0            100   \n",
       "\n",
       "                           variable     value  \\\n",
       "0  KT(RANK(DIV(WR)),RANK(DIV(COP)))  1.000000   \n",
       "1  KT(RANK(DIV(WR)),RANK(DIV(COP)))  1.000000   \n",
       "2  KT(RANK(DIV(WR)),RANK(DIV(COP)))  0.600000   \n",
       "3  KT(RANK(DIV(WR)),RANK(DIV(COP)))  0.333333   \n",
       "4  KT(RANK(DIV(WR)),RANK(DIV(COP)))  0.047619   \n",
       "\n",
       "                                          Comparison  \n",
       "0  KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}...  \n",
       "1  KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}...  \n",
       "2  KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}...  \n",
       "3  KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}...  \n",
       "4  KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}...  "
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Putting data in the format of the image above\n",
    "subp = df.melt(id_vars=['number_issues','number_profiles','number_agents'],\\\n",
    "                 value_vars=[\"KT(RANK(DIV(WR)),RANK(DIV(COP)))\",\n",
    "                             \"KT(RANK(DIV(WR)),STD)\",\n",
    "                             \"KT(RANK(DIV(COP)),STD)\"])\n",
    "\n",
    "subp['Comparison']= None\n",
    "subp.loc[subp['variable']=='KT(RANK(DIV(WR)),RANK(DIV(COP)))','Comparison']= \"KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Div}^\\mathit{Copeland}$)\"\n",
    "subp.loc[subp['variable']=='KT(RANK(DIV(WR)),STD)','Comparison']= \"KT($\\mathit{Div}^\\mathit{Borda}$,$\\mathit{Variance}$)\"\n",
    "subp.loc[subp['variable']=='KT(RANK(DIV(COP)),STD)', 'Comparison']= \"KT($\\mathit{Div}^\\mathit{Copeland}$,$\\mathit{Variance}$)\"\n",
    "subp.head()\n",
    "\n",
    "subp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "0325d592",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Figure1: kendall-tau on the vertical axis, \n",
    "# number of individuals on horiz axis (fix number of items=20)\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(4,4), dpi=150)\n",
    "\n",
    "g = sns.lineplot(data=subp, \\\n",
    "             x='number_issues', \n",
    "                 y='value',\\\n",
    "                 hue='Comparison',\n",
    "             linewidth=2, \n",
    "                 style='Comparison',\\\n",
    "                 markers = True,\n",
    "                 alpha=0.8, \n",
    "                 palette='Dark2', legend=True,\n",
    "              ax=ax)\n",
    "leg = g.legend(title=\"\", frameon=False, fontsize=6, loc='upper right')\n",
    "leg.get_title().set_position((-10, 0))\n",
    "for line in leg.get_lines():\n",
    "    line.set_linewidth(2)\n",
    "\n",
    "\n",
    "ax.spines[\"left\"].set_linewidth(1.2)\n",
    "ax.spines[\"bottom\"].set_linewidth(1.2)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.set_ylim(-0.45,1.05)\n",
    "plt.title(\"UM10\")\n",
    "import matplotlib.ticker as mticker\n",
    "ax.xaxis.set_major_locator(mticker.MultipleLocator(3))\n",
    "ax.set_ylabel(\"KT\")\n",
    "ax.set_xlabel(\"# issues\")\n",
    "plt.tight_layout()\n",
    "# plt.savefig(\"kt_div_between_agents.svg\")\n",
    "plt.savefig(\"figures/figure_experiment_1.pdf\")\n",
    "# ax[0].legend(title='Data',bbox_to_anchor=(1.2,0.9))\n",
    "# ax[0].legend(title='Data',bbox_to_anchor=(1.2,0.9))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c1036b34",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "748b777c",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}