{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "706e2fc4",
   "metadata": {},
   "source": [
    "# Withdrawal prediction on behavioral indicators\n",
    "\n",
    "In this section, we try to reproduce the withdrawal prediction approach based on\n",
    "behavioral indicators and Synthetic Minority Over-sampling (SMOTE) as presented in\n",
    "the work of Hlioui et al. {cite}`hlioui_2021`.\n",
    "\n",
    "Given the recurring occurrence of elevated learner withdrawal rates in Massive Open\n",
    "Online Courses (MOOCs), the implementation of early withdrawal prediction models\n",
    "could facilitate pedagogical enhancements, enable tailored intervention strategies\n",
    "and empower learners to monitor and enhance their academic performance.\n",
    "\n",
    "The approach of Hlioui et al. encloses four main phases:\n",
    "1. Data preprocessing\n",
    "2. Behavioral indicators extraction (feature engineering)\n",
    "3. K-means-based Data Discretizing\n",
    "4. Withdrawal prediction\n",
    "\n",
    "```{bibliography}\n",
    ":filter: docname in docnames\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "cbd1cf6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "from itertools import chain\n",
    "\n",
    "import pandas as pd\n",
    "from imblearn.over_sampling import SMOTE\n",
    "from IPython.display import display\n",
    "from matplotlib import pyplot as plt\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import GridSearchCV, StratifiedKFold, train_test_split\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.preprocessing import MinMaxScaler, OrdinalEncoder\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "from oulad import filter_by_module_presentation, get_oulad\n",
    "\n",
    "%load_ext oulad.capture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4b9ddb85",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture oulad\n",
    "oulad = get_oulad()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31ed3580",
   "metadata": {},
   "source": [
    "## Data preprocessing\n",
    "\n",
    "As in the work of Hlioui et al., we use the data from the `DDD` course `2013B`\n",
    "presentation.\n",
    "We start by extracting three tables containing data related to student demographics,\n",
    "student assessments and student course interactions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e99b9ca4",
   "metadata": {},
   "outputs": [],
   "source": [
    "MODULE = \"DDD\"\n",
    "PRESENTATION = \"2013B\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "684541f6",
   "metadata": {},
   "source": [
    "### Student demographics table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "123cbd84",
   "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>id_student</th>\n",
       "      <th>gender</th>\n",
       "      <th>region</th>\n",
       "      <th>highest_education</th>\n",
       "      <th>age_band</th>\n",
       "      <th>num_of_prev_attempts</th>\n",
       "      <th>disability</th>\n",
       "      <th>final_result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>13091</th>\n",
       "      <td>24213</td>\n",
       "      <td>F</td>\n",
       "      <td>East Anglian Region</td>\n",
       "      <td>A Level or Equivalent</td>\n",
       "      <td>0-35</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>Withdrawn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13092</th>\n",
       "      <td>40419</td>\n",
       "      <td>M</td>\n",
       "      <td>East Midlands Region</td>\n",
       "      <td>Lower Than A Level</td>\n",
       "      <td>35-55</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>Withdrawn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13093</th>\n",
       "      <td>41060</td>\n",
       "      <td>M</td>\n",
       "      <td>Ireland</td>\n",
       "      <td>HE Qualification</td>\n",
       "      <td>0-35</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>Fail</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13094</th>\n",
       "      <td>43284</td>\n",
       "      <td>M</td>\n",
       "      <td>East Midlands Region</td>\n",
       "      <td>Lower Than A Level</td>\n",
       "      <td>0-35</td>\n",
       "      <td>2</td>\n",
       "      <td>N</td>\n",
       "      <td>Withdrawn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13095</th>\n",
       "      <td>45664</td>\n",
       "      <td>M</td>\n",
       "      <td>Yorkshire Region</td>\n",
       "      <td>HE Qualification</td>\n",
       "      <td>0-35</td>\n",
       "      <td>1</td>\n",
       "      <td>N</td>\n",
       "      <td>Pass</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14389</th>\n",
       "      <td>2693243</td>\n",
       "      <td>F</td>\n",
       "      <td>Yorkshire Region</td>\n",
       "      <td>Lower Than A Level</td>\n",
       "      <td>35-55</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>Distinction</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14390</th>\n",
       "      <td>2694933</td>\n",
       "      <td>F</td>\n",
       "      <td>East Midlands Region</td>\n",
       "      <td>A Level or Equivalent</td>\n",
       "      <td>0-35</td>\n",
       "      <td>0</td>\n",
       "      <td>Y</td>\n",
       "      <td>Pass</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14391</th>\n",
       "      <td>2697773</td>\n",
       "      <td>F</td>\n",
       "      <td>East Anglian Region</td>\n",
       "      <td>A Level or Equivalent</td>\n",
       "      <td>0-35</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>Fail</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14392</th>\n",
       "      <td>2707979</td>\n",
       "      <td>F</td>\n",
       "      <td>East Midlands Region</td>\n",
       "      <td>Lower Than A Level</td>\n",
       "      <td>0-35</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>Fail</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14393</th>\n",
       "      <td>2710343</td>\n",
       "      <td>M</td>\n",
       "      <td>North Western Region</td>\n",
       "      <td>Lower Than A Level</td>\n",
       "      <td>0-35</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>Fail</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1303 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       id_student gender                region      highest_education  \\\n",
       "13091       24213      F   East Anglian Region  A Level or Equivalent   \n",
       "13092       40419      M  East Midlands Region     Lower Than A Level   \n",
       "13093       41060      M               Ireland       HE Qualification   \n",
       "13094       43284      M  East Midlands Region     Lower Than A Level   \n",
       "13095       45664      M      Yorkshire Region       HE Qualification   \n",
       "...           ...    ...                   ...                    ...   \n",
       "14389     2693243      F      Yorkshire Region     Lower Than A Level   \n",
       "14390     2694933      F  East Midlands Region  A Level or Equivalent   \n",
       "14391     2697773      F   East Anglian Region  A Level or Equivalent   \n",
       "14392     2707979      F  East Midlands Region     Lower Than A Level   \n",
       "14393     2710343      M  North Western Region     Lower Than A Level   \n",
       "\n",
       "      age_band  num_of_prev_attempts disability final_result  \n",
       "13091     0-35                     0          N    Withdrawn  \n",
       "13092    35-55                     0          N    Withdrawn  \n",
       "13093     0-35                     0          N         Fail  \n",
       "13094     0-35                     2          N    Withdrawn  \n",
       "13095     0-35                     1          N         Pass  \n",
       "...        ...                   ...        ...          ...  \n",
       "14389    35-55                     0          N  Distinction  \n",
       "14390     0-35                     0          Y         Pass  \n",
       "14391     0-35                     0          N         Fail  \n",
       "14392     0-35                     0          N         Fail  \n",
       "14393     0-35                     0          N         Fail  \n",
       "\n",
       "[1303 rows x 8 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "student_info = filter_by_module_presentation(\n",
    "    oulad.student_info, MODULE, PRESENTATION\n",
    ").drop(columns=[\"studied_credits\", \"imd_band\"])\n",
    "display(student_info)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "63fde837",
   "metadata": {},
   "source": [
    "### Student assessments table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "900088db",
   "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>date</th>\n",
       "      <th>weight</th>\n",
       "      <th>id_student</th>\n",
       "      <th>date_submitted</th>\n",
       "      <th>score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>23.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>907261</td>\n",
       "      <td>25</td>\n",
       "      <td>80.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>23.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>957467</td>\n",
       "      <td>25</td>\n",
       "      <td>77.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>23.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1018685</td>\n",
       "      <td>25</td>\n",
       "      <td>40.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1044992</td>\n",
       "      <td>25</td>\n",
       "      <td>73.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>23.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1068316</td>\n",
       "      <td>25</td>\n",
       "      <td>67.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10368</th>\n",
       "      <td>240.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2677955</td>\n",
       "      <td>230</td>\n",
       "      <td>69.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10369</th>\n",
       "      <td>240.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2683836</td>\n",
       "      <td>230</td>\n",
       "      <td>71.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10370</th>\n",
       "      <td>240.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2689536</td>\n",
       "      <td>229</td>\n",
       "      <td>58.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10371</th>\n",
       "      <td>240.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2693243</td>\n",
       "      <td>230</td>\n",
       "      <td>82.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10372</th>\n",
       "      <td>240.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2694933</td>\n",
       "      <td>230</td>\n",
       "      <td>73.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10373 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        date  weight  id_student  date_submitted  score\n",
       "0       23.0     2.0      907261              25   80.0\n",
       "1       23.0     2.0      957467              25   77.0\n",
       "2       23.0     2.0     1018685              25   40.0\n",
       "3       23.0     2.0     1044992              25   73.0\n",
       "4       23.0     2.0     1068316              25   67.0\n",
       "...      ...     ...         ...             ...    ...\n",
       "10368  240.0   100.0     2677955             230   69.0\n",
       "10369  240.0   100.0     2683836             230   71.0\n",
       "10370  240.0   100.0     2689536             229   58.0\n",
       "10371  240.0   100.0     2693243             230   82.0\n",
       "10372  240.0   100.0     2694933             230   73.0\n",
       "\n",
       "[10373 rows x 5 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "assessments = filter_by_module_presentation(oulad.assessments, MODULE, PRESENTATION)\n",
    "student_assessments = assessments.merge(\n",
    "    oulad.student_assessment, on=\"id_assessment\"\n",
    ").drop(columns=[\"is_banked\", \"assessment_type\", \"id_assessment\"])\n",
    "display(student_assessments)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0c1a609",
   "metadata": {},
   "source": [
    "### Student course interactions table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "adece966",
   "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>id_student</th>\n",
       "      <th>sum_click</th>\n",
       "      <th>activity_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>430516</td>\n",
       "      <td>3</td>\n",
       "      <td>page</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>430516</td>\n",
       "      <td>6</td>\n",
       "      <td>homepage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>430516</td>\n",
       "      <td>2</td>\n",
       "      <td>url</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>420388</td>\n",
       "      <td>1</td>\n",
       "      <td>url</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>420388</td>\n",
       "      <td>3</td>\n",
       "      <td>homepage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>536832</th>\n",
       "      <td>536170</td>\n",
       "      <td>2</td>\n",
       "      <td>resource</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>536833</th>\n",
       "      <td>536926</td>\n",
       "      <td>1</td>\n",
       "      <td>homepage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>536834</th>\n",
       "      <td>556295</td>\n",
       "      <td>3</td>\n",
       "      <td>homepage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>536835</th>\n",
       "      <td>556780</td>\n",
       "      <td>1</td>\n",
       "      <td>homepage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>536836</th>\n",
       "      <td>556918</td>\n",
       "      <td>2</td>\n",
       "      <td>homepage</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>536837 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        id_student  sum_click activity_type\n",
       "0           430516          3          page\n",
       "1           430516          6      homepage\n",
       "2           430516          2           url\n",
       "3           420388          1           url\n",
       "4           420388          3      homepage\n",
       "...            ...        ...           ...\n",
       "536832      536170          2      resource\n",
       "536833      536926          1      homepage\n",
       "536834      556295          3      homepage\n",
       "536835      556780          1      homepage\n",
       "536836      556918          2      homepage\n",
       "\n",
       "[536837 rows x 3 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%capture -ns withdrawal_prediction_on_behavioral_indicators student_vle\n",
    "student_vle = (\n",
    "    filter_by_module_presentation(oulad.student_vle, MODULE, PRESENTATION)\n",
    "    .merge(filter_by_module_presentation(oulad.vle, MODULE, PRESENTATION), on=\"id_site\")\n",
    "    .drop(columns=[\"id_site\", \"date\", \"week_from\", \"week_to\"])\n",
    ")\n",
    "display(student_vle)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "002f5f38",
   "metadata": {},
   "source": [
    "### Distribution of learners' results in the course DDD2013B\n",
    "\n",
    "Next, we reproduce the pie chart from the paper of Hlioui et al. (Figure 3)\n",
    "which reveals class imbalance of student final outcome in the `DDD2013B` course\n",
    "presentation.\n",
    "\n",
    "The class imbalance problem is a commonly recognized challenge in Machine Learning,\n",
    "known to hinder the development of effective classifiers. (Batista, et al., 2004)\n",
    "\n",
    "When trained on imbalanced datasets, models often exhibit a significant bias toward\n",
    "the majority class.\n",
    "This bias occurs because traditional Machine Learning algorithms are primarily\n",
    "focused on maximizing overall prediction accuracy, which can lead them to overlook or\n",
    "neglect classes with fewer instances, typically referred to as the minority class.\n",
    "(Bekkar & Alitouche, 2013)\n",
    "\n",
    "To deal with the class imbalance problem, the approach of Hlioui et al. applies the\n",
    "Synthetic Minority Over-sampling method (SMOTE) on the dataset, which generates new\n",
    "observations in the minority class by interpolating the existing ones.\n",
    "\n",
    "```{note}\n",
    "At this juncture, we deviate marginally from the initial approach as we apply the\n",
    "SMOTE method subsequent to the extraction of behavioral indicators.\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "852df139",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "(\n",
    "    student_info.final_result.value_counts()\n",
    "    .loc[[\"Withdrawn\", \"Fail\", \"Pass\", \"Distinction\"]]\n",
    "    .plot.pie(\n",
    "        title=\"Distribution of learners' results in the DDD2013B course\",\n",
    "        ylabel=\"\",\n",
    "        wedgeprops={\"width\": 0.6, \"edgecolor\": \"w\"},\n",
    "        autopct=\"%1.0f%%\",\n",
    "        pctdistance=0.72,\n",
    "        colors=[\"firebrick\", \"steelblue\", \"silver\", \"darkkhaki\"],\n",
    "        startangle=-270,\n",
    "        counterclock=False,\n",
    "        labeldistance=None,\n",
    "        radius=1.2,\n",
    "        textprops={\"color\": \"white\", \"weight\": \"bold\", \"fontsize\": 12.5},\n",
    "    )\n",
    ")\n",
    "plt.legend(loc=\"center\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74b51cfc",
   "metadata": {},
   "source": [
    "## Behavioral indicators extraction\n",
    "\n",
    "At this stage we extract behavioral indicators as described in the work of Hlioui et\n",
    "al.\n",
    "\n",
    "### Autonomy\n",
    "\n",
    "The autonomy indicator refers to the navigation frequency of learners within the\n",
    "virtual learning environment (VLE)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c3110ca3",
   "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>autonomy</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id_student</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40419</th>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41060</th>\n",
       "      <td>214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43284</th>\n",
       "      <td>422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45664</th>\n",
       "      <td>405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52014</th>\n",
       "      <td>116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2691780</th>\n",
       "      <td>156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2692101</th>\n",
       "      <td>151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2693243</th>\n",
       "      <td>1249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2694933</th>\n",
       "      <td>398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2697773</th>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1214 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            autonomy\n",
       "id_student          \n",
       "40419             60\n",
       "41060            214\n",
       "43284            422\n",
       "45664            405\n",
       "52014            116\n",
       "...              ...\n",
       "2691780          156\n",
       "2692101          151\n",
       "2693243         1249\n",
       "2694933          398\n",
       "2697773           86\n",
       "\n",
       "[1214 rows x 1 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "autonomy = (\n",
    "    student_vle.drop(columns=\"activity_type\")\n",
    "    .groupby([\"id_student\"])\n",
    "    .count()\n",
    "    .rename(columns={\"sum_click\": \"autonomy\"})\n",
    ")\n",
    "display(autonomy)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11a098c6",
   "metadata": {},
   "source": [
    "### Perseverance\n",
    "\n",
    "The perseverance indicator refers to the ratio of evaluations submitted on time by\n",
    "the learners."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "ba28759e",
   "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>perseverance</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id_student</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40419</th>\n",
       "      <td>0.071429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41060</th>\n",
       "      <td>0.357143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43284</th>\n",
       "      <td>0.214286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45664</th>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52014</th>\n",
       "      <td>0.357143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2689536</th>\n",
       "      <td>0.285714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2691780</th>\n",
       "      <td>0.214286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2692101</th>\n",
       "      <td>0.071429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2693243</th>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2694933</th>\n",
       "      <td>0.428571</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>969 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            perseverance\n",
       "id_student              \n",
       "40419           0.071429\n",
       "41060           0.357143\n",
       "43284           0.214286\n",
       "45664           0.500000\n",
       "52014           0.357143\n",
       "...                  ...\n",
       "2689536         0.285714\n",
       "2691780         0.214286\n",
       "2692101         0.071429\n",
       "2693243         0.500000\n",
       "2694933         0.428571\n",
       "\n",
       "[969 rows x 1 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "perseverance = (\n",
    "    student_assessments.query(\"date_submitted <= date\")[[\"id_student\", \"score\"]]\n",
    "    .groupby(\"id_student\")\n",
    "    .count()\n",
    "    .div(assessments.shape[0])\n",
    "    .rename(columns={\"score\": \"perseverance\"})\n",
    ")\n",
    "display(perseverance)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "480d0786",
   "metadata": {},
   "source": [
    "### Commitment indicators\n",
    "\n",
    "Commitment indicators aim to measure the level and type of involvement of learners.\n",
    "In the work of Hlioui et al. they refer to the total sum of clicks (interactions)\n",
    "made by learners on several related activity types (activity categories).\n",
    "\n",
    "Below, a barplot is presented, illustrating the frequency distribution of\n",
    "interactions by activity type within the DDD2013B course."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "6366aa9e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "student_vle.activity_type.value_counts().to_frame().plot.barh(\n",
    "    title=\"Activity type frequency in the DDD2013B course\", xlabel=\"Frequency\"\n",
    ")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "edd63309",
   "metadata": {},
   "source": [
    "### Collaborative commitment\n",
    "\n",
    "The collaborative commitment indicator refers to the sum of clicks learners made on\n",
    "activities of type `forumng`, `ouwiki`, and `ouelluminate`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0a3f681b",
   "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>collaborative_commitment</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id_student</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40419</th>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41060</th>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43284</th>\n",
       "      <td>176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45664</th>\n",
       "      <td>181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52014</th>\n",
       "      <td>110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2691780</th>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2692101</th>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2693243</th>\n",
       "      <td>941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2694933</th>\n",
       "      <td>305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2697773</th>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1106 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            collaborative_commitment\n",
       "id_student                          \n",
       "40419                             69\n",
       "41060                             46\n",
       "43284                            176\n",
       "45664                            181\n",
       "52014                            110\n",
       "...                              ...\n",
       "2691780                           43\n",
       "2692101                           32\n",
       "2693243                          941\n",
       "2694933                          305\n",
       "2697773                           32\n",
       "\n",
       "[1106 rows x 1 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "collaborative_commitment = (\n",
    "    student_vle.query(\"activity_type in ['forumng', 'ouwiki', 'ouelluminate']\")\n",
    "    .drop(columns=\"activity_type\")\n",
    "    .groupby(\"id_student\")\n",
    "    .sum()\n",
    "    .rename(columns={\"sum_click\": \"collaborative_commitment\"})\n",
    ")\n",
    "display(collaborative_commitment)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7bfd6c39",
   "metadata": {},
   "source": [
    "### Course structure commitment\n",
    "\n",
    "The course structure commitment indicator refers to the sum of clicks learners made on\n",
    "activities of type `homepage` and `glossary`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "86797910",
   "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>course_structure_commitment</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id_student</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40419</th>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41060</th>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43284</th>\n",
       "      <td>293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45664</th>\n",
       "      <td>287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52014</th>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2691780</th>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2692101</th>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2693243</th>\n",
       "      <td>591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2694933</th>\n",
       "      <td>279</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2697773</th>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1211 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            course_structure_commitment\n",
       "id_student                             \n",
       "40419                                29\n",
       "41060                                80\n",
       "43284                               293\n",
       "45664                               287\n",
       "52014                                98\n",
       "...                                 ...\n",
       "2691780                              87\n",
       "2692101                              69\n",
       "2693243                             591\n",
       "2694933                             279\n",
       "2697773                              37\n",
       "\n",
       "[1211 rows x 1 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "course_structure_commitment = (\n",
    "    student_vle.query(\"activity_type in ['homepage', 'glossary']\")\n",
    "    .drop(columns=\"activity_type\")\n",
    "    .groupby(\"id_student\")\n",
    "    .sum()\n",
    "    .rename(columns={\"sum_click\": \"course_structure_commitment\"})\n",
    ")\n",
    "display(course_structure_commitment)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "87aef27b",
   "metadata": {},
   "source": [
    "### Course content commitment\n",
    "\n",
    "The course content commitment indicator refers to the sum of clicks learners made on\n",
    "activities of type `resource`, `url`, `oucontent`, `page`, and `subpage`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "579a7742",
   "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>course_content_commitment</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id_student</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40419</th>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41060</th>\n",
       "      <td>270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43284</th>\n",
       "      <td>348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45664</th>\n",
       "      <td>558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52014</th>\n",
       "      <td>132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2691780</th>\n",
       "      <td>154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2692101</th>\n",
       "      <td>215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2693243</th>\n",
       "      <td>1050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2694933</th>\n",
       "      <td>310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2697773</th>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1206 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            course_content_commitment\n",
       "id_student                           \n",
       "40419                              45\n",
       "41060                             270\n",
       "43284                             348\n",
       "45664                             558\n",
       "52014                             132\n",
       "...                               ...\n",
       "2691780                           154\n",
       "2692101                           215\n",
       "2693243                          1050\n",
       "2694933                           310\n",
       "2697773                            79\n",
       "\n",
       "[1206 rows x 1 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "course_content_commitment = (\n",
    "    student_vle.query(\n",
    "        \"activity_type in ['resource', 'url', 'oucontent', 'page', 'subpage']\"\n",
    "    )\n",
    "    .drop(columns=\"activity_type\")\n",
    "    .groupby(\"id_student\")\n",
    "    .sum()\n",
    "    .rename(columns={\"sum_click\": \"course_content_commitment\"})\n",
    ")\n",
    "display(course_content_commitment)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f0972ae",
   "metadata": {},
   "source": [
    "### Evaluation activities commitment\n",
    "\n",
    "The evaluation activities commitment indicator refers to the sum of clicks learners\n",
    "made on activities of type `extenalquiz`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "1458f198",
   "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>evalutation_activities_commitment</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id_student</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40419</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41060</th>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43284</th>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45664</th>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52014</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2689536</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2691780</th>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2692101</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2693243</th>\n",
       "      <td>61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2694933</th>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1103 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            evalutation_activities_commitment\n",
       "id_student                                   \n",
       "40419                                       1\n",
       "41060                                      15\n",
       "43284                                      30\n",
       "45664                                      22\n",
       "52014                                       9\n",
       "...                                       ...\n",
       "2689536                                     8\n",
       "2691780                                    24\n",
       "2692101                                     2\n",
       "2693243                                    61\n",
       "2694933                                    41\n",
       "\n",
       "[1103 rows x 1 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "evalutation_activities_commitment = (\n",
    "    student_vle.query(\"activity_type == 'externalquiz'\")\n",
    "    .drop(columns=\"activity_type\")\n",
    "    .groupby(\"id_student\")\n",
    "    .sum()\n",
    "    .rename(columns={\"sum_click\": \"evalutation_activities_commitment\"})\n",
    ")\n",
    "display(evalutation_activities_commitment)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e3dca3c",
   "metadata": {},
   "source": [
    "### Motivation\n",
    "\n",
    "The motivation indicator measures whether a learners' sum of clicks on all activities\n",
    "is above average (motivated) or below (unmotivated)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "67cc79e2",
   "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>motivation</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id_student</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40419</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41060</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43284</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45664</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52014</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2691780</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2692101</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2693243</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2694933</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2697773</th>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1214 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            motivation\n",
       "id_student            \n",
       "40419              0.0\n",
       "41060              0.0\n",
       "43284              0.0\n",
       "45664              0.0\n",
       "52014              0.0\n",
       "...                ...\n",
       "2691780            0.0\n",
       "2692101            0.0\n",
       "2693243            1.0\n",
       "2694933            0.0\n",
       "2697773            0.0\n",
       "\n",
       "[1214 rows x 1 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "motivation = (\n",
    "    student_vle.drop(columns=\"activity_type\")\n",
    "    .groupby(\"id_student\")\n",
    "    .sum()\n",
    "    .assign(motivation=lambda df: df.sum_click >= df.sum_click.mean())\n",
    "    .drop(columns=\"sum_click\")\n",
    "    .astype(float)\n",
    ")\n",
    "display(motivation)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2fc23d11",
   "metadata": {},
   "source": [
    "## Performance\n",
    "\n",
    "The performance indicator refers to the sum of weighted assessment scores by learner."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "647b99ea",
   "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>performance</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id_student</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40419</th>\n",
       "      <td>704.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41060</th>\n",
       "      <td>6359.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43284</th>\n",
       "      <td>4020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45664</th>\n",
       "      <td>9528.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52014</th>\n",
       "      <td>4424.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2689536</th>\n",
       "      <td>10621.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2691780</th>\n",
       "      <td>2703.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2692101</th>\n",
       "      <td>1819.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2693243</th>\n",
       "      <td>17058.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2694933</th>\n",
       "      <td>14997.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1065 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            performance\n",
       "id_student             \n",
       "40419             704.0\n",
       "41060            6359.5\n",
       "43284            4020.0\n",
       "45664            9528.0\n",
       "52014            4424.0\n",
       "...                 ...\n",
       "2689536         10621.0\n",
       "2691780          2703.0\n",
       "2692101          1819.0\n",
       "2693243         17058.0\n",
       "2694933         14997.0\n",
       "\n",
       "[1065 rows x 1 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "performance = (\n",
    "    student_assessments.assign(performance=lambda df: df.weight * df.score)\n",
    "    .drop(columns=[\"date\", \"date_submitted\", \"weight\", \"score\"])\n",
    "    .groupby(\"id_student\")\n",
    "    .sum()\n",
    ")\n",
    "display(performance)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21be0370",
   "metadata": {},
   "source": [
    "## K-Means-based data discretizing\n",
    "\n",
    "At this stage, we discretize the generated indicators using the K-Means clustering\n",
    "method.\n",
    "We begin by estimating the 'k' parameter through the elbow method and then replace\n",
    "the indicators' values with their corresponding clustering labels.\n",
    "\n",
    "### Elbow method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f2bcf392",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x2000 with 7 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%capture -ns withdrawal_prediction_on_behavioral_indicators inertia\n",
    "indicators = {\n",
    "    \"Perseverance indicator\": perseverance,\n",
    "    \"Autonomy indicator\": autonomy,\n",
    "    \"Structure content commitment indicator\": course_structure_commitment,\n",
    "    \"Evaluation activities commitment indicator\": evalutation_activities_commitment,\n",
    "    \"Learning content commitment indicator\": course_content_commitment,\n",
    "    \"Collaborative commitment indicator\": collaborative_commitment,\n",
    "    \"Performance indicator\": performance,\n",
    "}\n",
    "k_range = list(range(2, 10))\n",
    "fig = plt.figure(figsize=(20, 20))\n",
    "# Inertia: Sum of squared distances of samples to their closest cluster center,\n",
    "# weighted by the sample weights if provided.\n",
    "inertia = [\n",
    "    KMeans(n_clusters=k, n_init=\"auto\").fit(indicator.values).inertia_\n",
    "    for indicator in indicators.values()\n",
    "    for k in k_range\n",
    "]\n",
    "for i, name in enumerate(indicators):\n",
    "    index = i * len(k_range)\n",
    "    pd.Series(inertia[index : index + len(k_range)], index=k_range, name=\"k\").plot(\n",
    "        title=f\"{name} (Inertia/k)\",\n",
    "        xlabel=\"Number of clusters\",\n",
    "        ylabel=\"SSE\",\n",
    "        grid=True,\n",
    "        marker=\"s\",\n",
    "        ax=plt.subplot2grid((5, 5), (int(i / 2), 2 * (i % 2) + int(i / 6)), colspan=2),\n",
    "    )\n",
    "\n",
    "plt.subplots_adjust(hspace=0.4, wspace=0.4)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "37bf1e54",
   "metadata": {},
   "source": [
    "### Discretization\n",
    "\n",
    "We obtain similar graphs as in the work of Hlioui et al.\n",
    "Thus, we proceed by setting the same `k` values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "e345a480",
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_k = {\n",
    "    \"Perseverance indicator\": 3,\n",
    "    \"Autonomy indicator\": 3,\n",
    "    \"Structure content commitment indicator\": 2,\n",
    "    \"Evaluation activities commitment indicator\": 3,\n",
    "    \"Learning content commitment indicator\": 2,\n",
    "    \"Collaborative commitment indicator\": 3,\n",
    "    \"Performance indicator\": 3,\n",
    "}\n",
    "for name, k in selected_k.items():\n",
    "    kmeans = KMeans(n_clusters=k, n_init=\"auto\")\n",
    "    indicator = indicators[name]\n",
    "    indicator[indicator.columns[0]] = kmeans.fit_predict(indicator)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24225362",
   "metadata": {},
   "source": [
    "### Feature table\n",
    "\n",
    "We join the student demographics table with the discretized behavioral\n",
    "indicators into a single `feature_table`.\n",
    "\n",
    "We also encode categorical columns (`age_band`, `disability`, `gender`,\n",
    "`highest_education`, `region`, and `final_result`) to numerical values and fill\n",
    "missing values with zeros."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "13a5cd09",
   "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>gender</th>\n",
       "      <th>region</th>\n",
       "      <th>highest_education</th>\n",
       "      <th>age_band</th>\n",
       "      <th>num_of_prev_attempts</th>\n",
       "      <th>disability</th>\n",
       "      <th>final_result</th>\n",
       "      <th>motivation</th>\n",
       "      <th>perseverance</th>\n",
       "      <th>autonomy</th>\n",
       "      <th>course_structure_commitment</th>\n",
       "      <th>evalutation_activities_commitment</th>\n",
       "      <th>course_content_commitment</th>\n",
       "      <th>collaborative_commitment</th>\n",
       "      <th>performance</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id_student</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>24213</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40419</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41060</th>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43284</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45664</th>\n",
       "      <td>0.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2693243</th>\n",
       "      <td>1.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2694933</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2697773</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2707979</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2710343</th>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1303 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            gender  region  highest_education  age_band  num_of_prev_attempts  \\\n",
       "id_student                                                                      \n",
       "24213          1.0     0.0               0.50       0.0                   0.0   \n",
       "40419          0.0     1.0               0.25       0.5                   0.0   \n",
       "41060          0.0     2.0               0.75       0.0                   0.0   \n",
       "43284          0.0     1.0               0.25       0.0                   2.0   \n",
       "45664          0.0    12.0               0.75       0.0                   1.0   \n",
       "...            ...     ...                ...       ...                   ...   \n",
       "2693243        1.0    12.0               0.25       0.5                   0.0   \n",
       "2694933        1.0     1.0               0.50       0.0                   0.0   \n",
       "2697773        1.0     0.0               0.50       0.0                   0.0   \n",
       "2707979        1.0     1.0               0.25       0.0                   0.0   \n",
       "2710343        0.0     5.0               0.25       0.0                   0.0   \n",
       "\n",
       "            disability  final_result  motivation  perseverance  autonomy  \\\n",
       "id_student                                                                 \n",
       "24213              0.0           1.0         0.0           0.0       0.0   \n",
       "40419              0.0           1.0         0.0           0.0       0.0   \n",
       "41060              0.0           0.0         0.0           2.0       0.0   \n",
       "43284              0.0           1.0         0.0           2.0       1.0   \n",
       "45664              0.0           0.0         0.0           1.0       1.0   \n",
       "...                ...           ...         ...           ...       ...   \n",
       "2693243            0.0           0.0         1.0           1.0       2.0   \n",
       "2694933            1.0           0.0         0.0           1.0       1.0   \n",
       "2697773            0.0           0.0         0.0           0.0       0.0   \n",
       "2707979            0.0           0.0         0.0           0.0       0.0   \n",
       "2710343            0.0           0.0         0.0           0.0       0.0   \n",
       "\n",
       "            course_structure_commitment  evalutation_activities_commitment  \\\n",
       "id_student                                                                   \n",
       "24213                               0.0                                0.0   \n",
       "40419                               1.0                                2.0   \n",
       "41060                               1.0                                2.0   \n",
       "43284                               1.0                                0.0   \n",
       "45664                               1.0                                0.0   \n",
       "...                                 ...                                ...   \n",
       "2693243                             1.0                                1.0   \n",
       "2694933                             1.0                                0.0   \n",
       "2697773                             1.0                                0.0   \n",
       "2707979                             0.0                                0.0   \n",
       "2710343                             0.0                                0.0   \n",
       "\n",
       "            course_content_commitment  collaborative_commitment  performance  \n",
       "id_student                                                                    \n",
       "24213                             0.0                       0.0          0.0  \n",
       "40419                             0.0                       1.0          1.0  \n",
       "41060                             0.0                       1.0          2.0  \n",
       "43284                             0.0                       1.0          1.0  \n",
       "45664                             0.0                       1.0          2.0  \n",
       "...                               ...                       ...          ...  \n",
       "2693243                           1.0                       0.0          0.0  \n",
       "2694933                           0.0                       1.0          0.0  \n",
       "2697773                           0.0                       1.0          0.0  \n",
       "2707979                           0.0                       0.0          0.0  \n",
       "2710343                           0.0                       0.0          0.0  \n",
       "\n",
       "[1303 rows x 15 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "region_encoder = OrdinalEncoder()\n",
    "feature_table = (\n",
    "    student_info.set_index(\"id_student\")\n",
    "    .join(chain([motivation], indicators.values()), how=\"outer\")\n",
    "    .fillna(0.0)\n",
    "    .replace(\n",
    "        {\n",
    "            \"age_band\": {\"0-35\": \"0.0\", \"35-55\": \"0.5\", \"55<=\": \"1.0\"},\n",
    "            \"disability\": {\"N\": \"0.0\", \"Y\": \"1.0\"},\n",
    "            \"gender\": {\"M\": \"0.0\", \"F\": \"1.0\"},\n",
    "            \"highest_education\": {\n",
    "                \"No Formal quals\": \"0.0\",\n",
    "                \"Lower Than A Level\": \"0.25\",\n",
    "                \"A Level or Equivalent\": \"0.5\",\n",
    "                \"HE Qualification\": \"0.75\",\n",
    "                \"Post Graduate Qualification\": \"1.0\",\n",
    "            },\n",
    "            \"final_result\": {\n",
    "                \"Withdrawn\": \"1.0\",\n",
    "                \"Fail\": \"0.0\",\n",
    "                \"Pass\": \"0.0\",\n",
    "                \"Distinction\": \"0.0\",\n",
    "            },\n",
    "        }\n",
    "    )\n",
    "    .assign(region=lambda df: region_encoder.fit_transform(df[[\"region\"]]))\n",
    "    .astype(float)\n",
    ")\n",
    "display(feature_table)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "506b9c6b",
   "metadata": {},
   "source": [
    "## Withdrawal prediction\n",
    "\n",
    "Prior to training the classification models, we split the `feature_table` into a\n",
    "train (75%) and test (25%) set.\n",
    "Then we scale features to values between 0 and 1 and apply the SMOTE method to\n",
    "balance the occurences of the target class (Withdrawn/Not Withdrawn).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "e2517847",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, x_test, y_train, y_test = train_test_split(\n",
    "    feature_table.drop(columns=\"final_result\").values, feature_table.final_result.values\n",
    ")\n",
    "scaler = MinMaxScaler()\n",
    "x_train = scaler.fit_transform(x_train)\n",
    "x_test = scaler.transform(x_test)\n",
    "smote = SMOTE()\n",
    "x_resampled, y_resampled = smote.fit_resample(x_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13c36f6b",
   "metadata": {},
   "source": [
    "Subsequently, we undertake a grid search across classification parameters for five\n",
    "classifiers, which have been chosen to align closely with those\n",
    "utilized in the work of Hlioui et al.\n",
    "These classifiers encompass Decision Trees, Random Forest, Support Vector Machines,\n",
    "Gaussian Naive Bayes (as a substitute for Tree Augmented Naive Bayes), and\n",
    "Multilayer Perceptron.\n",
    "\n",
    "In line with the work of Hlioui et al., we adopt a 5-fold cross-validation approach\n",
    "with stratification.\n",
    "Evaluation of classifier performance is conducted utilizing the F-measure as the\n",
    "primary metric.\n",
    "\n",
    "```{note}\n",
    "We replaced the initial parameter ranges with the selected values after the first\n",
    "grid search run to speed up the process.\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "9d34492c",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">cv_score</th>\n",
       "      <th colspan=\"2\" halign=\"left\">test_score</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dataset</th>\n",
       "      <th>balanced</th>\n",
       "      <th>unbalanced</th>\n",
       "      <th>balanced</th>\n",
       "      <th>unbalanced</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>classifier</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DecisionTreeClassifier</th>\n",
       "      <td>0.839965</td>\n",
       "      <td>0.705460</td>\n",
       "      <td>0.740458</td>\n",
       "      <td>0.726531</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GaussianNB</th>\n",
       "      <td>0.800166</td>\n",
       "      <td>0.690658</td>\n",
       "      <td>0.683453</td>\n",
       "      <td>0.693431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MLPClassifier</th>\n",
       "      <td>0.833515</td>\n",
       "      <td>0.708588</td>\n",
       "      <td>0.794007</td>\n",
       "      <td>0.737288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RandomForestClassifier</th>\n",
       "      <td>0.850020</td>\n",
       "      <td>0.659298</td>\n",
       "      <td>0.733871</td>\n",
       "      <td>0.681416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC</th>\n",
       "      <td>0.853157</td>\n",
       "      <td>0.698794</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.726531</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        cv_score            test_score           \n",
       "dataset                 balanced unbalanced   balanced unbalanced\n",
       "classifier                                                       \n",
       "DecisionTreeClassifier  0.839965   0.705460   0.740458   0.726531\n",
       "GaussianNB              0.800166   0.690658   0.683453   0.693431\n",
       "MLPClassifier           0.833515   0.708588   0.794007   0.737288\n",
       "RandomForestClassifier  0.850020   0.659298   0.733871   0.681416\n",
       "SVC                     0.853157   0.698794   0.750000   0.726531"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%capture -ns withdrawal_prediction_on_behavioral_indicators scores\n",
    "grid = {\n",
    "    DecisionTreeClassifier: {\n",
    "        \"unbalanced\": {\n",
    "            \"criterion\": [\"log_loss\"],  # [\"gini\", \"entropy\", \"log_loss\"],\n",
    "            \"max_depth\": [3],  # [None, *list(range(1, 20))],\n",
    "            \"min_samples_leaf\": [13],  # range(1, 20),\n",
    "            \"min_samples_split\": [9],  # range(2, 20),\n",
    "            \"splitter\": [\"random\"],  # [\"random\", \"best\"],\n",
    "        },\n",
    "        \"balanced\": {\n",
    "            \"criterion\": [\"entropy\"],  # [\"gini\", \"entropy\", \"log_loss\"],\n",
    "            \"max_depth\": [8],  # [None, *list(range(1, 20))],\n",
    "            \"min_samples_leaf\": [1],  # range(1, 20),\n",
    "            \"min_samples_split\": [12],  # range(2, 20),\n",
    "            \"splitter\": [\"random\"],  # [\"random\", \"best\"],\n",
    "        },\n",
    "    },\n",
    "    RandomForestClassifier: {\n",
    "        \"unbalanced\": {\n",
    "            \"criterion\": [\"entropy\"],  # [\"gini\", \"entropy\", \"log_loss\"],\n",
    "            \"max_depth\": [15],  # [None, *list(range(1, 20, 2))],\n",
    "            \"min_samples_leaf\": [3],  # list(range(1, 20, 2)),\n",
    "            \"min_samples_split\": [2],  # list(range(2, 20, 2)),\n",
    "            \"n_estimators\": [10],  # [10, 50, 100],\n",
    "        },\n",
    "        \"balanced\": {\n",
    "            \"criterion\": [\"entropy\"],  # [\"gini\", \"entropy\", \"log_loss\"],\n",
    "            \"max_depth\": [17],  # [None, *list(range(1, 20, 2))],\n",
    "            \"min_samples_leaf\": [3],  # list(range(1, 20, 2)),\n",
    "            \"min_samples_split\": [14],  # list(range(2, 20, 2)),\n",
    "            \"n_estimators\": [50],  # [10, 50, 100],\n",
    "        },\n",
    "    },\n",
    "    GaussianNB: {\n",
    "        \"unbalanced\": {\n",
    "            \"var_smoothing\": [0.001],  # [1/10**x for x in range(1, 11)],\n",
    "        },\n",
    "        \"balanced\": {\n",
    "            \"var_smoothing\": [0.001],  # [1/10**x for x in range(1, 11)],\n",
    "        },\n",
    "    },\n",
    "    SVC: {\n",
    "        \"unbalanced\": {\n",
    "            \"C\": [1.0],  # [1.0],\n",
    "            \"gamma\": [0.5],  # [\"scale\", \"auto\", 0, 0.5],\n",
    "            \"kernel\": [\"poly\"],  # [\"rbf\", \"poly\", \"sigmoid\"],\n",
    "            \"tol\": [0.001],  # [1/10**x for x in range(2, 5)],\n",
    "        },\n",
    "        \"balanced\": {\n",
    "            \"C\": [1.0],  # [1.0],\n",
    "            \"gamma\": [0.5],  # [\"scale\", \"auto\", 0, 0.5],\n",
    "            \"kernel\": [\"poly\"],  # [\"rbf\", \"poly\", \"sigmoid\"],\n",
    "            \"tol\": [0.01],  # [1/10**x for x in range(2, 5)],\n",
    "        },\n",
    "    },\n",
    "    MLPClassifier: {\n",
    "        \"unbalanced\": {\n",
    "            \"early_stopping\": [False],  # [True, False],\n",
    "            \"hidden_layer_sizes\": [(10, 10, 500)],  # 1-3 layers of 10/100/500 nodes\n",
    "            \"learning_rate\": [\"constant\"],\n",
    "            \"learning_rate_init\": [0.3],  # [0.001, 0.1, 0.3],\n",
    "            \"max_iter\": [1200],\n",
    "            \"momentum\": [0.9],  # [0.2, 0.5, 0.9],\n",
    "            \"solver\": [\"sgd\"],\n",
    "        },\n",
    "        \"balanced\": {\n",
    "            \"early_stopping\": [False],  # [True, False],\n",
    "            \"hidden_layer_sizes\": [(100, 500)],  # 1-3 layers of 10/100/500 nodes\n",
    "            \"learning_rate\": [\"constant\"],\n",
    "            \"learning_rate_init\": [0.1],  # [0.001, 0.1, 0.3],\n",
    "            \"max_iter\": [1200],\n",
    "            \"momentum\": [0.5],  # [0.2, 0.5, 0.9],\n",
    "            \"solver\": [\"sgd\"],\n",
    "        },\n",
    "    },\n",
    "}\n",
    "\n",
    "\n",
    "def get_scores():\n",
    "    \"\"\"Yields scores for each classifier from the grid.\"\"\"\n",
    "    skf_cv = StratifiedKFold(n_splits=5, shuffle=True)\n",
    "    for classifier_class, type_parameters in grid.items():\n",
    "        for dataset_type, hyperparameters in type_parameters.items():\n",
    "            classifier = GridSearchCV(\n",
    "                classifier_class(),\n",
    "                hyperparameters,\n",
    "                scoring=\"f1\",\n",
    "                n_jobs=-1,\n",
    "                error_score=\"raise\",\n",
    "                cv=skf_cv,\n",
    "                refit=True,\n",
    "            )\n",
    "            if dataset_type == \"unbalanced\":\n",
    "                classifier.fit(x_train, y_train)\n",
    "            else:\n",
    "                classifier.fit(x_resampled, y_resampled)\n",
    "            cv_score = classifier.best_score_\n",
    "            test_score = classifier.score(x_test, y_test)\n",
    "            yield (classifier_class.__name__, dataset_type, cv_score, test_score)\n",
    "\n",
    "\n",
    "scores = list(get_scores())\n",
    "display(\n",
    "    pd.DataFrame(\n",
    "        scores, columns=[\"classifier\", \"dataset\", \"cv_score\", \"test_score\"]\n",
    "    ).pivot_table(\n",
    "        values=[\"cv_score\", \"test_score\"], index=[\"classifier\"], columns=[\"dataset\"]\n",
    "    )\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "jupytext": {
   "cell_metadata_filter": "-all",
   "main_language": "python",
   "notebook_metadata_filter": "-all"
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}