{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "958832c8",
      "metadata": {
        "id": "958832c8"
      },
      "source": [
        "\n",
        "# ExtraaLearn Lead Conversion Prediction — Full End‑to‑End Notebook\n",
        "\n",
        "This notebook performs:\n",
        "- Data Loading & Overview\n",
        "- Exploratory Data Analysis (EDA)\n",
        "- Preprocessing Pipeline (ColumnTransformer)\n",
        "- Model Building (Random Forest, Gradient Boosting)\n",
        "- Hyperparameter Tuning\n",
        "- Model Evaluation (ROC‑AUC, PR‑AUC, F1, Recall)\n",
        "- Model Serialization\n",
        "- Actionable Insights and Business Recommendations\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fd0GZeIVSQPl",
        "outputId": "1126552d-2f0c-42d1-cc40-a3cc42aac49d"
      },
      "id": "fd0GZeIVSQPl",
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%cd /content/drive/MyDrive/extraLearn"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qX2Ay0JVSQXM",
        "outputId": "cf10e880-aaeb-430e-ebb0-86c0ae9e6a58"
      },
      "id": "qX2Ay0JVSQXM",
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/drive/MyDrive/extraLearn\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "id": "1220f5bd",
      "metadata": {
        "id": "1220f5bd"
      },
      "outputs": [],
      "source": [
        "\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "from pathlib import Path\n",
        "from sklearn.model_selection import train_test_split, StratifiedKFold, RandomizedSearchCV\n",
        "from sklearn.metrics import roc_auc_score, average_precision_score, f1_score, recall_score, precision_score, confusion_matrix, classification_report\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n",
        "from sklearn.compose import ColumnTransformer\n",
        "from sklearn.preprocessing import OneHotEncoder, OrdinalEncoder, StandardScaler\n",
        "from sklearn.impute import SimpleImputer\n",
        "import joblib\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "b28eb868",
      "metadata": {
        "id": "b28eb868"
      },
      "source": [
        "## Load Dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "id": "005ffcb7",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 226
        },
        "id": "005ffcb7",
        "outputId": "ee1c3121-f945-429a-b823-36b359896b56"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       ID  age current_occupation first_interaction profile_completed  \\\n",
              "0  EXT001   57         Unemployed           Website              High   \n",
              "1  EXT002   56       Professional        Mobile App            Medium   \n",
              "2  EXT003   52       Professional           Website            Medium   \n",
              "3  EXT004   53         Unemployed           Website              High   \n",
              "4  EXT005   23            Student           Website              High   \n",
              "\n",
              "   website_visits  time_spent_on_website  page_views_per_visit  \\\n",
              "0               7                   1639                 1.861   \n",
              "1               2                     83                 0.320   \n",
              "2               3                    330                 0.074   \n",
              "3               4                    464                 2.057   \n",
              "4               4                    600                16.914   \n",
              "\n",
              "      last_activity print_media_type1 print_media_type2 digital_media  \\\n",
              "0  Website Activity               Yes                No           Yes   \n",
              "1  Website Activity                No                No            No   \n",
              "2  Website Activity                No                No           Yes   \n",
              "3  Website Activity                No                No            No   \n",
              "4    Email Activity                No                No            No   \n",
              "\n",
              "  educational_channels referral  status  \n",
              "0                   No       No       1  \n",
              "1                  Yes       No       0  \n",
              "2                   No       No       0  \n",
              "3                   No       No       1  \n",
              "4                   No       No       0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-4332968e-f4e8-4554-b8ba-3703eedf530c\" class=\"colab-df-container\">\n",
              "    <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</th>\n",
              "      <th>age</th>\n",
              "      <th>current_occupation</th>\n",
              "      <th>first_interaction</th>\n",
              "      <th>profile_completed</th>\n",
              "      <th>website_visits</th>\n",
              "      <th>time_spent_on_website</th>\n",
              "      <th>page_views_per_visit</th>\n",
              "      <th>last_activity</th>\n",
              "      <th>print_media_type1</th>\n",
              "      <th>print_media_type2</th>\n",
              "      <th>digital_media</th>\n",
              "      <th>educational_channels</th>\n",
              "      <th>referral</th>\n",
              "      <th>status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>EXT001</td>\n",
              "      <td>57</td>\n",
              "      <td>Unemployed</td>\n",
              "      <td>Website</td>\n",
              "      <td>High</td>\n",
              "      <td>7</td>\n",
              "      <td>1639</td>\n",
              "      <td>1.861</td>\n",
              "      <td>Website Activity</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>EXT002</td>\n",
              "      <td>56</td>\n",
              "      <td>Professional</td>\n",
              "      <td>Mobile App</td>\n",
              "      <td>Medium</td>\n",
              "      <td>2</td>\n",
              "      <td>83</td>\n",
              "      <td>0.320</td>\n",
              "      <td>Website Activity</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>EXT003</td>\n",
              "      <td>52</td>\n",
              "      <td>Professional</td>\n",
              "      <td>Website</td>\n",
              "      <td>Medium</td>\n",
              "      <td>3</td>\n",
              "      <td>330</td>\n",
              "      <td>0.074</td>\n",
              "      <td>Website Activity</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>EXT004</td>\n",
              "      <td>53</td>\n",
              "      <td>Unemployed</td>\n",
              "      <td>Website</td>\n",
              "      <td>High</td>\n",
              "      <td>4</td>\n",
              "      <td>464</td>\n",
              "      <td>2.057</td>\n",
              "      <td>Website Activity</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>EXT005</td>\n",
              "      <td>23</td>\n",
              "      <td>Student</td>\n",
              "      <td>Website</td>\n",
              "      <td>High</td>\n",
              "      <td>4</td>\n",
              "      <td>600</td>\n",
              "      <td>16.914</td>\n",
              "      <td>Email Activity</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4332968e-f4e8-4554-b8ba-3703eedf530c')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-4332968e-f4e8-4554-b8ba-3703eedf530c button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-4332968e-f4e8-4554-b8ba-3703eedf530c');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "df",
              "summary": "{\n  \"name\": \"df\",\n  \"rows\": 4612,\n  \"fields\": [\n    {\n      \"column\": \"ID\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4612,\n        \"samples\": [\n          \"EXT1074\",\n          \"EXT2750\",\n          \"EXT1375\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"age\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 13,\n        \"min\": 18,\n        \"max\": 63,\n        \"num_unique_values\": 46,\n        \"samples\": [\n          33,\n          19,\n          28\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"current_occupation\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Unemployed\",\n          \"Professional\",\n          \"Student\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"first_interaction\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Mobile App\",\n          \"Website\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"profile_completed\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"High\",\n          \"Medium\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"website_visits\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2,\n        \"min\": 0,\n        \"max\": 30,\n        \"num_unique_values\": 27,\n        \"samples\": [\n          12,\n          14\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"time_spent_on_website\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 743,\n        \"min\": 0,\n        \"max\": 2537,\n        \"num_unique_values\": 1623,\n        \"samples\": [\n          2509,\n          1051\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"page_views_per_visit\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.9681246688859402,\n        \"min\": 0.0,\n        \"max\": 18.434,\n        \"num_unique_values\": 2414,\n        \"samples\": [\n          2.745,\n          1.93\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"last_activity\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Website Activity\",\n          \"Email Activity\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"print_media_type1\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"No\",\n          \"Yes\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"print_media_type2\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Yes\",\n          \"No\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"digital_media\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"No\",\n          \"Yes\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"educational_channels\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Yes\",\n          \"No\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"referral\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Yes\",\n          \"No\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"status\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 0,\n        \"max\": 1,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0,\n          1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 20
        }
      ],
      "source": [
        "\n",
        "df = pd.read_csv('ExtraaLearn.csv')\n",
        "df.head()\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "89e23876",
      "metadata": {
        "id": "89e23876"
      },
      "source": [
        "## Basic Data Cleaning"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "id": "090488de",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 796
        },
        "id": "090488de",
        "outputId": "d8c866df-686d-4e4b-8322-6afd8c657df2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 4612 entries, 0 to 4611\n",
            "Data columns (total 15 columns):\n",
            " #   Column                 Non-Null Count  Dtype  \n",
            "---  ------                 --------------  -----  \n",
            " 0   ID                     4612 non-null   object \n",
            " 1   age                    4612 non-null   int64  \n",
            " 2   current_occupation     4612 non-null   object \n",
            " 3   first_interaction      4612 non-null   object \n",
            " 4   profile_completed      4612 non-null   object \n",
            " 5   website_visits         4612 non-null   int64  \n",
            " 6   time_spent_on_website  4612 non-null   int64  \n",
            " 7   page_views_per_visit   4612 non-null   float64\n",
            " 8   last_activity          4612 non-null   object \n",
            " 9   print_media_type1      4612 non-null   object \n",
            " 10  print_media_type2      4612 non-null   object \n",
            " 11  digital_media          4612 non-null   object \n",
            " 12  educational_channels   4612 non-null   object \n",
            " 13  referral               4612 non-null   object \n",
            " 14  status                 4612 non-null   int64  \n",
            "dtypes: float64(1), int64(4), object(10)\n",
            "memory usage: 540.6+ KB\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "             ID          age current_occupation first_interaction  \\\n",
              "count      4612  4612.000000               4612              4612   \n",
              "unique     4612          NaN                  3                 2   \n",
              "top     EXT4612          NaN       Professional           Website   \n",
              "freq          1          NaN               2616              2542   \n",
              "mean        NaN    46.201214                NaN               NaN   \n",
              "std         NaN    13.161454                NaN               NaN   \n",
              "min         NaN    18.000000                NaN               NaN   \n",
              "25%         NaN    36.000000                NaN               NaN   \n",
              "50%         NaN    51.000000                NaN               NaN   \n",
              "75%         NaN    57.000000                NaN               NaN   \n",
              "max         NaN    63.000000                NaN               NaN   \n",
              "\n",
              "       profile_completed  website_visits  time_spent_on_website  \\\n",
              "count               4612     4612.000000            4612.000000   \n",
              "unique                 3             NaN                    NaN   \n",
              "top       High (75-100%)             NaN                    NaN   \n",
              "freq                2264             NaN                    NaN   \n",
              "mean                 NaN        3.566782             724.011275   \n",
              "std                  NaN        2.829134             743.828683   \n",
              "min                  NaN        0.000000               0.000000   \n",
              "25%                  NaN        2.000000             148.750000   \n",
              "50%                  NaN        3.000000             376.000000   \n",
              "75%                  NaN        5.000000            1336.750000   \n",
              "max                  NaN       30.000000            2537.000000   \n",
              "\n",
              "        page_views_per_visit   last_activity print_media_type1  \\\n",
              "count            4612.000000            4612              4612   \n",
              "unique                   NaN               3                 2   \n",
              "top                      NaN  Email Activity                No   \n",
              "freq                     NaN            2278              4115   \n",
              "mean                3.026126             NaN               NaN   \n",
              "std                 1.968125             NaN               NaN   \n",
              "min                 0.000000             NaN               NaN   \n",
              "25%                 2.077750             NaN               NaN   \n",
              "50%                 2.792000             NaN               NaN   \n",
              "75%                 3.756250             NaN               NaN   \n",
              "max                18.434000             NaN               NaN   \n",
              "\n",
              "       print_media_type2 digital_media educational_channels referral  \\\n",
              "count               4612          4612                 4612     4612   \n",
              "unique                 2             2                    2        2   \n",
              "top                   No            No                   No       No   \n",
              "freq                4379          4085                 3907     4519   \n",
              "mean                 NaN           NaN                  NaN      NaN   \n",
              "std                  NaN           NaN                  NaN      NaN   \n",
              "min                  NaN           NaN                  NaN      NaN   \n",
              "25%                  NaN           NaN                  NaN      NaN   \n",
              "50%                  NaN           NaN                  NaN      NaN   \n",
              "75%                  NaN           NaN                  NaN      NaN   \n",
              "max                  NaN           NaN                  NaN      NaN   \n",
              "\n",
              "             status  \n",
              "count   4612.000000  \n",
              "unique          NaN  \n",
              "top             NaN  \n",
              "freq            NaN  \n",
              "mean       0.298569  \n",
              "std        0.457680  \n",
              "min        0.000000  \n",
              "25%        0.000000  \n",
              "50%        0.000000  \n",
              "75%        1.000000  \n",
              "max        1.000000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-64c6965c-fcf3-448a-93c0-b5c94287ce71\" class=\"colab-df-container\">\n",
              "    <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</th>\n",
              "      <th>age</th>\n",
              "      <th>current_occupation</th>\n",
              "      <th>first_interaction</th>\n",
              "      <th>profile_completed</th>\n",
              "      <th>website_visits</th>\n",
              "      <th>time_spent_on_website</th>\n",
              "      <th>page_views_per_visit</th>\n",
              "      <th>last_activity</th>\n",
              "      <th>print_media_type1</th>\n",
              "      <th>print_media_type2</th>\n",
              "      <th>digital_media</th>\n",
              "      <th>educational_channels</th>\n",
              "      <th>referral</th>\n",
              "      <th>status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>4612</td>\n",
              "      <td>4612.000000</td>\n",
              "      <td>4612</td>\n",
              "      <td>4612</td>\n",
              "      <td>4612</td>\n",
              "      <td>4612.000000</td>\n",
              "      <td>4612.000000</td>\n",
              "      <td>4612.000000</td>\n",
              "      <td>4612</td>\n",
              "      <td>4612</td>\n",
              "      <td>4612</td>\n",
              "      <td>4612</td>\n",
              "      <td>4612</td>\n",
              "      <td>4612</td>\n",
              "      <td>4612.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>unique</th>\n",
              "      <td>4612</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>top</th>\n",
              "      <td>EXT4612</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Professional</td>\n",
              "      <td>Website</td>\n",
              "      <td>High (75-100%)</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Email Activity</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>freq</th>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2616</td>\n",
              "      <td>2542</td>\n",
              "      <td>2264</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2278</td>\n",
              "      <td>4115</td>\n",
              "      <td>4379</td>\n",
              "      <td>4085</td>\n",
              "      <td>3907</td>\n",
              "      <td>4519</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>NaN</td>\n",
              "      <td>46.201214</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3.566782</td>\n",
              "      <td>724.011275</td>\n",
              "      <td>3.026126</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.298569</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>NaN</td>\n",
              "      <td>13.161454</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2.829134</td>\n",
              "      <td>743.828683</td>\n",
              "      <td>1.968125</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.457680</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>NaN</td>\n",
              "      <td>18.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>NaN</td>\n",
              "      <td>36.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>148.750000</td>\n",
              "      <td>2.077750</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>NaN</td>\n",
              "      <td>51.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>376.000000</td>\n",
              "      <td>2.792000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>NaN</td>\n",
              "      <td>57.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>1336.750000</td>\n",
              "      <td>3.756250</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>NaN</td>\n",
              "      <td>63.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>30.000000</td>\n",
              "      <td>2537.000000</td>\n",
              "      <td>18.434000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-64c6965c-fcf3-448a-93c0-b5c94287ce71')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-64c6965c-fcf3-448a-93c0-b5c94287ce71 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-64c6965c-fcf3-448a-93c0-b5c94287ce71');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"df\",\n  \"rows\": 11,\n  \"fields\": [\n    {\n      \"column\": \"ID\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"4612\",\n          \"EXT4612\",\n          \"1\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"age\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1616.3227600990067,\n        \"min\": 13.161454018142754,\n        \"max\": 4612.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          46.20121422376409,\n          51.0,\n          4612.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"current_occupation\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          3,\n          \"2616\",\n          \"4612\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"first_interaction\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          2,\n          \"2542\",\n          \"4612\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"profile_completed\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          3,\n          \"2264\",\n          \"4612\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"website_visits\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1628.2734572990112,\n        \"min\": 0.0,\n        \"max\": 4612.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          3.566782307025152,\n          3.0,\n          4612.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"time_spent_on_website\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1558.8612385248775,\n        \"min\": 0.0,\n        \"max\": 4612.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          724.0112749349523,\n          376.0,\n          4612.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"page_views_per_visit\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1628.9794386217416,\n        \"min\": 0.0,\n        \"max\": 4612.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          3.0261255420641806,\n          2.792,\n          4612.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"last_activity\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          3,\n          \"2278\",\n          \"4612\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"print_media_type1\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          2,\n          \"4115\",\n          \"4612\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"print_media_type2\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          2,\n          \"4379\",\n          \"4612\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"digital_media\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          2,\n          \"4085\",\n          \"4612\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"educational_channels\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          2,\n          \"3907\",\n          \"4612\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"referral\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          2,\n          \"4519\",\n          \"4612\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"status\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1630.4490789788301,\n        \"min\": 0.0,\n        \"max\": 4612.0,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.29856895056374677,\n          1.0,\n          0.4576799656995658\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 21
        }
      ],
      "source": [
        "\n",
        "# Normalize ordinal labels\n",
        "if 'profile_completed' in df.columns:\n",
        "    df['profile_completed'] = df['profile_completed'].replace({\n",
        "        'Low': 'Low (0-50%)',\n",
        "        'Medium': 'Medium (50-75%)',\n",
        "        'High': 'High (75-100%)'\n",
        "    })\n",
        "\n",
        "# Display summary\n",
        "df.info()\n",
        "df.describe(include='all')\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "07c8e5d7",
      "metadata": {
        "id": "07c8e5d7"
      },
      "source": [
        "## Exploratory Data Analysis (EDA)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "id": "5f830855",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "5f830855",
        "outputId": "67f61e30-8ae5-4faf-9154-f01fea8d513e"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x300 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x300 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x300 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x300 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "\n",
        "# Distribution of target\n",
        "sns.countplot(data=df, x='status')\n",
        "plt.title('Target Distribution')\n",
        "plt.show()\n",
        "\n",
        "# Numeric distributions\n",
        "num_cols = ['age','website_visits','time_spent_on_website','page_views_per_visit']\n",
        "df[num_cols].hist(figsize=(10,6))\n",
        "plt.show()\n",
        "\n",
        "# Conversion rate by categorical vars\n",
        "cat_cols = ['current_occupation','first_interaction','last_activity','profile_completed']\n",
        "for col in cat_cols:\n",
        "    rate = df.groupby(col)['status'].mean().sort_values(ascending=False)\n",
        "    rate.plot(kind='bar', figsize=(6,3), title=f'Conversion Rate by {col}')\n",
        "    plt.show()\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "c2857a4b",
      "metadata": {
        "id": "c2857a4b"
      },
      "source": [
        "## Preprocessing Pipeline"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "id": "b94741ce",
      "metadata": {
        "id": "b94741ce"
      },
      "outputs": [],
      "source": [
        "\n",
        "NUM_COLS = ['age','website_visits','time_spent_on_website','page_views_per_visit']\n",
        "BIN_COLS = ['print_media_type1','print_media_type2','digital_media','educational_channels','referral']\n",
        "CAT_COLS = ['current_occupation','first_interaction','last_activity']\n",
        "ORD_COLS = ['profile_completed']\n",
        "TARGET = 'status'\n",
        "\n",
        "ordinal_map = [['Low (0-50%)','Medium (50-75%)','High (75-100%)']]\n",
        "\n",
        "num_pipe = Pipeline([\n",
        "    ('imp', SimpleImputer(strategy='median')),\n",
        "    ('scaler', StandardScaler())\n",
        "])\n",
        "\n",
        "bin_pipe = Pipeline([\n",
        "    ('imp', SimpleImputer(strategy='most_frequent')),\n",
        "    ('enc', OrdinalEncoder(categories=[['No','Yes']]*len(BIN_COLS)))\n",
        "])\n",
        "\n",
        "cat_pipe = Pipeline([\n",
        "    ('imp', SimpleImputer(strategy='most_frequent')),\n",
        "    ('ohe', OneHotEncoder(handle_unknown='ignore'))\n",
        "])\n",
        "\n",
        "ord_pipe = Pipeline([\n",
        "    ('imp', SimpleImputer(strategy='most_frequent')),\n",
        "    ('ord', OrdinalEncoder(categories=ordinal_map))\n",
        "])\n",
        "\n",
        "preprocess = ColumnTransformer([\n",
        "    ('num', num_pipe, NUM_COLS),\n",
        "    ('bin', bin_pipe, BIN_COLS),\n",
        "    ('cat', cat_pipe, CAT_COLS),\n",
        "    ('ord', ord_pipe, ORD_COLS)\n",
        "])\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "3f9a1ae4",
      "metadata": {
        "id": "3f9a1ae4"
      },
      "source": [
        "## Train-Test Split"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "id": "8bcceb92",
      "metadata": {
        "id": "8bcceb92"
      },
      "outputs": [],
      "source": [
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "X = df[NUM_COLS + BIN_COLS + CAT_COLS + ORD_COLS]\n",
        "y = df[TARGET]\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, stratify=y, random_state=42)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "4db84f88",
      "metadata": {
        "id": "4db84f88"
      },
      "source": [
        "## Baseline Model Training"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "id": "9090eb9e",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9090eb9e",
        "outputId": "660395b0-2278-4b0a-a0d2-9d41b0351d9d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'RandomForest': np.float64(0.9144631480533882),\n",
              " 'GradientBoosting': np.float64(0.9228610159849652)}"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ],
      "source": [
        "\n",
        "cv = StratifiedKFold(n_splits=3, shuffle=True, random_state=42)\n",
        "\n",
        "rf = Pipeline([\n",
        "    ('pre', preprocess),\n",
        "    ('clf', RandomForestClassifier(n_estimators=200, random_state=42))\n",
        "])\n",
        "\n",
        "gb = Pipeline([\n",
        "    ('pre', preprocess),\n",
        "    ('clf', GradientBoostingClassifier(random_state=42))\n",
        "])\n",
        "\n",
        "models = {'RandomForest': rf, 'GradientBoosting': gb}\n",
        "cv_scores = {}\n",
        "\n",
        "for name, model in models.items():\n",
        "    scores = []\n",
        "    for tr, va in cv.split(X_train, y_train):\n",
        "        model.fit(X_train.iloc[tr], y_train.iloc[tr])\n",
        "        proba = model.predict_proba(X_train.iloc[va])[:,1]\n",
        "        scores.append(roc_auc_score(y_train.iloc[va], proba))\n",
        "    cv_scores[name] = np.mean(scores)\n",
        "\n",
        "cv_scores\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "59c281fc",
      "metadata": {
        "id": "59c281fc"
      },
      "source": [
        "## Hyperparameter Tuning"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "id": "e24367f0",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "e24367f0",
        "outputId": "7168bcd8-8c28-436d-a795-32d750f13c59"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'GradientBoosting'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 26
        }
      ],
      "source": [
        "\n",
        "best_model_name = max(cv_scores, key=cv_scores.get)\n",
        "best_model_name\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "id": "0d1df6f2",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0d1df6f2",
        "outputId": "258b3287-f8f2-4997-df20-5cbeb423f11c"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'clf__subsample': 1.0,\n",
              " 'clf__n_estimators': 150,\n",
              " 'clf__max_depth': 3,\n",
              " 'clf__learning_rate': 0.05}"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ],
      "source": [
        "\n",
        "if best_model_name == 'RandomForest':\n",
        "    base = rf\n",
        "    params = {\n",
        "        'clf__n_estimators': [150,200,300],\n",
        "        'clf__max_depth': [None,6,10],\n",
        "        'clf__min_samples_split': [2,5],\n",
        "        'clf__min_samples_leaf': [1,2]\n",
        "    }\n",
        "else:\n",
        "    base = gb\n",
        "    params = {\n",
        "        'clf__n_estimators': [100,150,200],\n",
        "        'clf__learning_rate': [0.03,0.05,0.1],\n",
        "        'clf__max_depth': [2,3],\n",
        "        'clf__subsample': [0.8,1.0]\n",
        "    }\n",
        "\n",
        "rs = RandomizedSearchCV(base, params, n_iter=10, scoring='roc_auc', cv=cv, n_jobs=-1, random_state=42)\n",
        "rs.fit(X_train, y_train)\n",
        "rs.best_params_\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "a30ab662",
      "metadata": {
        "id": "a30ab662"
      },
      "source": [
        "## Final Model Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "id": "955ca581",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "955ca581",
        "outputId": "5af6ac76-b230-42bd-f215-705bb905a025"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'roc_auc': np.float64(0.9278862307640614),\n",
              " 'pr_auc': np.float64(0.8383560457393716),\n",
              " 'f1': 0.7680890538033395,\n",
              " 'recall': 0.75,\n",
              " 'precision': 0.7870722433460076,\n",
              " 'confusion_matrix': array([[591,  56],\n",
              "        [ 69, 207]])}"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ],
      "source": [
        "\n",
        "best_model = rs.best_estimator_\n",
        "proba = best_model.predict_proba(X_test)[:,1]\n",
        "pred = (proba >= 0.5).astype(int)\n",
        "\n",
        "eval_metrics = {\n",
        "    'roc_auc': roc_auc_score(y_test, proba),\n",
        "    'pr_auc': average_precision_score(y_test, proba),\n",
        "    'f1': f1_score(y_test, pred),\n",
        "    'recall': recall_score(y_test, pred),\n",
        "    'precision': precision_score(y_test, pred),\n",
        "    'confusion_matrix': confusion_matrix(y_test, pred)\n",
        "}\n",
        "eval_metrics\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "152e56db",
      "metadata": {
        "id": "152e56db"
      },
      "source": [
        "## Serialize Final Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "id": "aae7f179",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aae7f179",
        "outputId": "08631d46-64c0-4bf5-9da2-c77c4b63b1fb"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['best_model.joblib']"
            ]
          },
          "metadata": {},
          "execution_count": 29
        }
      ],
      "source": [
        "\n",
        "import joblib\n",
        "joblib.dump(best_model, 'best_model.joblib')\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 🔍 Actionable Insights & Business Recommendations\n",
        "*(Based on EDA findings and final model performance)*\n",
        "\n",
        "---\n",
        "\n",
        "## **📌 I. Key Data‑Driven Insights**\n",
        "\n",
        "### **1. Profile Completion Is the Strongest Conversion Signal**\n",
        "- Conversion increases sharply with profile completion:  \n",
        "  - **High (75–100%) → 41.8%**  \n",
        "  - **Medium (50–75%) → 18.9%**  \n",
        "  - **Low (0–50%) → 7.5%**  \n",
        "- This indicates that *the more a lead invests in completing their profile, the stronger their purchase intent*.\n",
        "\n",
        "---\n",
        "\n",
        "### **2. Website-Based First Interaction Converts Far Better Than Mobile App**\n",
        "- **Website leads convert at 45.6%**, while  \n",
        "- **Mobile App leads convert at only 10.5%**  \n",
        "This shows that website traffic represents *higher-quality, high‑intent leads* compared to app-origin leads.\n",
        "\n",
        "---\n",
        "\n",
        "### **3. Last Interaction Type Strongly Predicts Lead Warmth**\n",
        "Conversion rates by last_activity:  \n",
        "- **Website Activity – 38.5%**  \n",
        "- **Email Activity – 30.3%**  \n",
        "- **Phone Activity – 21.3%**  \n",
        "\n",
        "Website & email interactions reflect **active interest**, while phone interactions lag behind.\n",
        "\n",
        "---\n",
        "\n",
        "### **4. Engagement Depth Influences Conversion**\n",
        "Correlation with target:  \n",
        "- **time_spent_on_website: +0.30** (strongest numeric driver)  \n",
        "- **age: +0.12**  \n",
        "- `website_visits` and `page_views_per_visit` show minimal correlation individually.\n",
        "\n",
        "Longer sessions suggest **research behavior** typical of high-intent users.\n",
        "\n",
        "---\n",
        "\n",
        "### **5. Occupation Impacts Conversion Probability**\n",
        "Conversion rates:  \n",
        "- **Professionals → 35.5%**  \n",
        "- **Unemployed → 26.6%**  \n",
        "- **Students → 11.7%**\n",
        "\n",
        "Professionals show **higher readiness and ability to pay**, making them a key target group.\n",
        "\n",
        "---\n",
        "\n",
        "## **📌 II. Model‑Based Insights**\n",
        "\n",
        "The final tuned model (Gradient Boosting Classifier) achieved:  \n",
        "- **ROC‑AUC:** 0.928  \n",
        "- **PR‑AUC:** 0.838  \n",
        "- **F1 Score:** 0.768  \n",
        "- **Recall:** 0.75  \n",
        "- **Precision:** 0.787  \n",
        "\n",
        "This confirms the model is **highly reliable** at identifying leads with strong conversion likelihood.\n",
        "\n",
        "---\n",
        "\n",
        "## **📌 III. Actionable Business Recommendations**\n",
        "\n",
        "### **1. Optimize the Profile Completion Journey**\n",
        "Since profile completion has the highest impact:\n",
        "- Add **progress bars**, nudges, tooltips, and reminders.  \n",
        "- Offer micro‑incentives (e.g., **free guides**, coupon codes).  \n",
        "- Trigger automated email sequences for low‑completion leads.\n",
        "\n",
        "---\n",
        "\n",
        "### **2. Focus Marketing Spend on Website Traffic**\n",
        "Because website-origin leads convert **4× better**:\n",
        "- Strengthen **SEO/SEM**, retargeting, and landing page optimization.  \n",
        "- Improve the mobile‑app onboarding experience to close the conversion gap.\n",
        "\n",
        "---\n",
        "\n",
        "### **3. Prioritize Email & Web-Active Leads in Sales Routing**\n",
        "Assign higher priority to:\n",
        "- Leads whose last activity was **Website** or **Email**.  \n",
        "- These users demonstrate significantly higher intent and should enter **high-touch workflows**.\n",
        "\n",
        "---\n",
        "\n",
        "### **4. Use Behavioral Signals for Personalized Outreach**\n",
        "- High `time_spent_on_website` → **Hot lead** → Immediate call/WhatsApp.  \n",
        "- Medium profile completion but low engagement → Target with content-rich emails.  \n",
        "- Low completion + low engagement → Add to long-term nurture pipelines.\n",
        "\n",
        "---\n",
        "\n",
        "### **5. Create Occupation-Based Lead Personas**\n",
        "- **Professionals:** Prioritize; highlight ROI, career growth, and certification value.  \n",
        "- **Students:** Focus on affordability, internships, and placement support.  \n",
        "- **Unemployed:** Offer flexible payment plans and reskilling pathways.\n",
        "\n",
        "---\n",
        "\n",
        "### **6. Deploy Probability-Based Lead Tiering**\n",
        "Using model-generated conversion probabilities:\n",
        "\n",
        "| Tier | Probability Range | Action |\n",
        "|------|------------------|--------|\n",
        "| **Hot Leads** | p ≥ 0.80 | Immediate sales call / WhatsApp follow-up |\n",
        "| **Warm Leads** | 0.50 ≤ p < 0.80 | Email + webinar invites + remarketing |\n",
        "| **Cold Leads** | p < 0.50 | Long-term nurture campaigns |\n",
        "\n",
        "This **maximizes sales efficiency** and reduces CAC.\n",
        "\n",
        "---\n",
        "\n",
        "### **7. Retrain & Monitor Model Performance Regularly**\n",
        "- Monitor drift in lead behavior (quarterly).  \n",
        "- Retrain model with new data for consistent performance.  \n",
        "- Track prediction outcomes to refine thresholds.\n",
        "\n",
        "---\n",
        "\n",
        "## **📌 IV. Strategic Takeaways**\n",
        "\n",
        "✔ *Profile completion, website entry, and deeper engagement consistently indicate high intent.*  \n",
        "✔ *Professionals and website-origin leads are the highest-value groups.*  \n",
        "✔ *Your model effectively identifies warm/hot leads, enabling smarter sales operations.*  \n",
        "✔ *Targeted interventions based on these insights can significantly boost overall conversion rates.*\n",
        "\n",
        "---"
      ],
      "metadata": {
        "id": "Dfif0NGQheTW"
      },
      "id": "Dfif0NGQheTW"
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 🔗 Hugging Face Docker Space URL for Backend Flask API's\n",
        "\n",
        "\n",
        "➡ **Hugging Face Backend Space URL:**  \n",
        "`https://huggingface.co/spaces/manoj112025/extraaLearnModel`\n",
        "\n"
      ],
      "metadata": {
        "id": "puLUsZbqAFlW"
      },
      "id": "puLUsZbqAFlW"
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 🔗 Hugging Face Docker Space URL for Streamlit ExtraLearn Model for frontend space\n",
        "\n",
        "\n",
        "➡ **Hugging Face Frontend Space URL:**  \n",
        "`https://huggingface.co/spaces/manoj112025/StreamLitExtraLearnFrontendModel`\n",
        "\n"
      ],
      "metadata": {
        "id": "-Qbk4hSkHK7G"
      },
      "id": "-Qbk4hSkHK7G"
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "JjKj3zsfEDaz"
      },
      "id": "JjKj3zsfEDaz",
      "execution_count": 29,
      "outputs": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}