{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XQZJSSopDet9",
        "outputId": "d7fe7f2e-b030-4b8d-e574-0a6e896e824d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: pandas in /usr/local/lib/python3.12/dist-packages (2.2.2)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.12/dist-packages (2.0.2)\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.12/dist-packages (3.10.0)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.12/dist-packages (1.6.1)\n",
            "Requirement already satisfied: statsmodels in /usr/local/lib/python3.12/dist-packages (0.14.6)\n",
            "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.12/dist-packages (from pandas) (2.9.0.post0)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.12/dist-packages (from pandas) (2025.2)\n",
            "Requirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.12/dist-packages (from pandas) (2026.1)\n",
            "Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.12/dist-packages (from matplotlib) (1.3.3)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.12/dist-packages (from matplotlib) (0.12.1)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.12/dist-packages (from matplotlib) (4.62.1)\n",
            "Requirement already satisfied: kiwisolver>=1.3.1 in /usr/local/lib/python3.12/dist-packages (from matplotlib) (1.5.0)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.12/dist-packages (from matplotlib) (26.1)\n",
            "Requirement already satisfied: pillow>=8 in /usr/local/lib/python3.12/dist-packages (from matplotlib) (11.3.0)\n",
            "Requirement already satisfied: pyparsing>=2.3.1 in /usr/local/lib/python3.12/dist-packages (from matplotlib) (3.3.2)\n",
            "Requirement already satisfied: scipy>=1.6.0 in /usr/local/lib/python3.12/dist-packages (from scikit-learn) (1.16.3)\n",
            "Requirement already satisfied: joblib>=1.2.0 in /usr/local/lib/python3.12/dist-packages (from scikit-learn) (1.5.3)\n",
            "Requirement already satisfied: threadpoolctl>=3.1.0 in /usr/local/lib/python3.12/dist-packages (from scikit-learn) (3.6.0)\n",
            "Requirement already satisfied: patsy>=0.5.6 in /usr/local/lib/python3.12/dist-packages (from statsmodels) (1.0.2)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.12/dist-packages (from python-dateutil>=2.8.2->pandas) (1.17.0)\n"
          ]
        }
      ],
      "source": [
        "!pip install pandas numpy matplotlib scikit-learn statsmodels"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import random\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "random.seed(42)\n",
        "np.random.seed(42)"
      ],
      "metadata": {
        "id": "yzXlo3jYDi0f"
      },
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"=\" * 50)\n",
        "print(\"STEP 1/2: Data Creation (real data + synthetic enrichment)\")\n",
        "print(\"=\" * 50)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EL_CQaWsDkGw",
        "outputId": "105e7fb3-54f6-4717-f474-86c61bdaf485"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "==================================================\n",
            "STEP 1/2: Data Creation (real data + synthetic enrichment)\n",
            "==================================================\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# ====================================================\n",
        "# GENERATE SYNTHETIC EU JOBS DATA\n",
        "# ====================================================\n",
        "import random\n",
        "from datetime import datetime, timedelta\n",
        "\n",
        "random.seed(42)\n",
        "\n",
        "DOMAINS = [\n",
        "    \"Budget and Finances\", \"European Policy\", \"Crisis management and Internal security\",\n",
        "    \"Law\", \"Defence\", \"Human Resources\", \"Information Technology\",\n",
        "    \"Justice and human rights\", \"Economics, Finance and Statistics\",\n",
        "    \"Transport\", \"Agriculture\", \"Environment, Climate change\", \"Public Health\",\n",
        "    \"Education and Culture\", \"External Relations\", \"Trade\", \"Audit\",\n",
        "    \"Communication\", \"Research and Innovation\", \"Statistics\", \"Energy\",\n",
        "    \"Migration\", \"Digital Affairs\", \"Tax and Customs\",\n",
        "]\n",
        "TITLES = [\n",
        "    \"Project Assistant\", \"Policy Officer\", \"Legal Officer\", \"Finance Agent\",\n",
        "    \"Communications Officer\", \"Cybersecurity Expert\", \"IT Officer\",\n",
        "    \"Senior Advisor\", \"Head of Unit\", \"Junior Analyst\", \"Senior Analyst\",\n",
        "    \"Programme Manager\", \"Auditor\", \"HR Officer\", \"Translator\",\n",
        "    \"Data Analyst\", \"Compliance Officer\", \"Procurement Officer\",\n",
        "    \"Research Officer\", \"Liaison Officer\", \"Statistician\", \"Economist\",\n",
        "    \"Risk Manager\", \"Operations Officer\", \"Coordinator\",\n",
        "]\n",
        "GRADES = [\"FG II\", \"FG III\", \"FG IV\", \"FG III, FG IV\", \"AD 5\", \"AD 6\", \"AD 7\", \"AD 8\", \"AD 10\", \"AD 12\", \"AD 14\"]\n",
        "GRADE_WEIGHTS = [3, 11, 17, 9, 4, 7, 5, 2, 2, 2, 3]\n",
        "CONTRACTS = [\"Contract staff\", \"Temporary staff\", \"Seconded national expert (SNE)\", \"Permanent official\"]\n",
        "CONTRACT_WEIGHTS = [46, 28, 24, 5]\n",
        "INSTITUTIONS = [\n",
        "    \"EU institutions\", \"(EDA) European Defence Agency\", \"(EUSPA) European Union Agency for the Space Programme\",\n",
        "    \"(FRONTEX) European Border and Coast Guard Agency\", \"(FRA) European Union Agency for Fundamental Rights\",\n",
        "    \"(EASA) European Union Aviation Safety Agency\", \"(Europol) European Union Agency for Law Enforcement Cooperation\",\n",
        "    \"(EFSA) European Food Safety Authority\", \"(ECB) European Central Bank\", \"(EMA) European Medicines Agency\",\n",
        "    \"(EUAA) European Union Agency for Asylum\", \"(EIB) European Investment Bank\", \"(EEA) European Environment Agency\",\n",
        "]\n",
        "LOCATIONS = [\n",
        "    \"Brussels (Belgium)\", \"Luxembourg (Luxembourg)\", \"The Hague (The Netherlands)\",\n",
        "    \"Warsaw (Poland)\", \"Frankfurt (Germany)\", \"Cologne (Germany)\", \"Parma (Italy)\",\n",
        "    \"Lisbon (Portugal)\", \"Paris (France)\", \"Vienna (Austria)\", \"Helsinki (Finland)\",\n",
        "    \"Madrid (Spain)\", \"Dublin (Ireland)\", \"Stockholm (Sweden)\",\n",
        "]\n",
        "\n",
        "synthetic_rows = []\n",
        "for i in range(200):\n",
        "    base = datetime(2026, 5, 1)\n",
        "    deadline = base + timedelta(days=random.randint(1, 180))\n",
        "    deadline_str = f\"{deadline.strftime('%d/%m/%Y')} - {random.choice(['13:00', '17:00', '23:59'])}\"\n",
        "    title = random.choice(TITLES)\n",
        "    inst = random.choice(INSTITUTIONS)\n",
        "    inst_short = inst.split(\")\")[0].replace(\"(\", \"\").lower() if \"(\" in inst else \"eu\"\n",
        "\n",
        "    synthetic_rows.append({\n",
        "        \"ID\": 19800 + i,\n",
        "        \"title\": title,\n",
        "        \"Domain(s)\": random.choice(DOMAINS),\n",
        "        \"Grade\": random.choices(GRADES, weights=GRADE_WEIGHTS, k=1)[0],\n",
        "        \"Type of contract\": random.choices(CONTRACTS, weights=CONTRACT_WEIGHTS, k=1)[0],\n",
        "        \"Institution(s)    \": inst,\n",
        "        \"Location(s)\": random.choice(LOCATIONS),\n",
        "        \"Deadline \": deadline_str,\n",
        "        \"Link to Content\": f\"https://eu-careers.europa.eu/en/job-opportunities/{title.lower().replace(' ', '-')}/{inst_short}-{19800+i}\",\n",
        "    })\n",
        "\n",
        "synthetic_df = pd.DataFrame(synthetic_rows)\n",
        "print(f\"Generated {len(synthetic_df)} synthetic job postings\")\n",
        "synthetic_df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 435
        },
        "id": "BQf2iEmtum_U",
        "outputId": "f01811d0-c4f5-435c-f86f-85c4f1c84bab"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Generated 200 synthetic job postings\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "      ID                title                          Domain(s)   Grade  \\\n",
              "0  19800    Project Assistant  Economics, Finance and Statistics   FG IV   \n",
              "1  19801  Procurement Officer            Research and Innovation   FG IV   \n",
              "2  19802      Liaison Officer                      Communication  FG III   \n",
              "3  19803           Translator  Economics, Finance and Statistics    AD 7   \n",
              "4  19804           HR Officer  Economics, Finance and Statistics  FG III   \n",
              "\n",
              "     Type of contract                            Institution(s)      \\\n",
              "0      Contract staff                (EIB) European Investment Bank   \n",
              "1      Contract staff                 (EDA) European Defence Agency   \n",
              "2     Temporary staff                               EU institutions   \n",
              "3      Contract staff               (EMA) European Medicines Agency   \n",
              "4  Permanent official  (EASA) European Union Aviation Safety Agency   \n",
              "\n",
              "               Location(s)           Deadline   \\\n",
              "0  Luxembourg (Luxembourg)  12/10/2026 - 13:00   \n",
              "1          Warsaw (Poland)  22/10/2026 - 23:59   \n",
              "2           Paris (France)  30/06/2026 - 23:59   \n",
              "3         Dublin (Ireland)  17/08/2026 - 13:00   \n",
              "4        Cologne (Germany)  11/06/2026 - 23:59   \n",
              "\n",
              "                                     Link to Content  \n",
              "0  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "1  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "2  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "3  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "4  https://eu-careers.europa.eu/en/job-opportunit...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-d8fbda0f-8ab7-47b8-839b-6774c3523b85\" 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>title</th>\n",
              "      <th>Domain(s)</th>\n",
              "      <th>Grade</th>\n",
              "      <th>Type of contract</th>\n",
              "      <th>Institution(s)</th>\n",
              "      <th>Location(s)</th>\n",
              "      <th>Deadline</th>\n",
              "      <th>Link to Content</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>19800</td>\n",
              "      <td>Project Assistant</td>\n",
              "      <td>Economics, Finance and Statistics</td>\n",
              "      <td>FG IV</td>\n",
              "      <td>Contract staff</td>\n",
              "      <td>(EIB) European Investment Bank</td>\n",
              "      <td>Luxembourg (Luxembourg)</td>\n",
              "      <td>12/10/2026 - 13:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>19801</td>\n",
              "      <td>Procurement Officer</td>\n",
              "      <td>Research and Innovation</td>\n",
              "      <td>FG IV</td>\n",
              "      <td>Contract staff</td>\n",
              "      <td>(EDA) European Defence Agency</td>\n",
              "      <td>Warsaw (Poland)</td>\n",
              "      <td>22/10/2026 - 23:59</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>19802</td>\n",
              "      <td>Liaison Officer</td>\n",
              "      <td>Communication</td>\n",
              "      <td>FG III</td>\n",
              "      <td>Temporary staff</td>\n",
              "      <td>EU institutions</td>\n",
              "      <td>Paris (France)</td>\n",
              "      <td>30/06/2026 - 23:59</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>19803</td>\n",
              "      <td>Translator</td>\n",
              "      <td>Economics, Finance and Statistics</td>\n",
              "      <td>AD 7</td>\n",
              "      <td>Contract staff</td>\n",
              "      <td>(EMA) European Medicines Agency</td>\n",
              "      <td>Dublin (Ireland)</td>\n",
              "      <td>17/08/2026 - 13:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>19804</td>\n",
              "      <td>HR Officer</td>\n",
              "      <td>Economics, Finance and Statistics</td>\n",
              "      <td>FG III</td>\n",
              "      <td>Permanent official</td>\n",
              "      <td>(EASA) European Union Aviation Safety Agency</td>\n",
              "      <td>Cologne (Germany)</td>\n",
              "      <td>11/06/2026 - 23:59</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</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-d8fbda0f-8ab7-47b8-839b-6774c3523b85')\"\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-d8fbda0f-8ab7-47b8-839b-6774c3523b85 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-d8fbda0f-8ab7-47b8-839b-6774c3523b85');\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": "synthetic_df",
              "summary": "{\n  \"name\": \"synthetic_df\",\n  \"rows\": 200,\n  \"fields\": [\n    {\n      \"column\": \"ID\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 57,\n        \"min\": 19800,\n        \"max\": 19999,\n        \"num_unique_values\": 200,\n        \"samples\": [\n          19895,\n          19815,\n          19830\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"title\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 25,\n        \"samples\": [\n          \"Statistician\",\n          \"IT Officer\",\n          \"Project Assistant\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Domain(s)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 24,\n        \"samples\": [\n          \"Public Health\",\n          \"Education and Culture\",\n          \"Economics, Finance and Statistics\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Grade\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 11,\n        \"samples\": [\n          \"FG III, FG IV\",\n          \"FG IV\",\n          \"FG II\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Type of contract\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"Temporary staff\",\n          \"Seconded national expert (SNE)\",\n          \"Contract staff\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Institution(s)    \",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 13,\n        \"samples\": [\n          \"(FRA) European Union Agency for Fundamental Rights\",\n          \"(EFSA) European Food Safety Authority\",\n          \"(EIB) European Investment Bank\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Location(s)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 14,\n        \"samples\": [\n          \"The Hague (The Netherlands)\",\n          \"Brussels (Belgium)\",\n          \"Luxembourg (Luxembourg)\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Deadline \",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 169,\n        \"samples\": [\n          \"31/07/2026 - 13:00\",\n          \"25/09/2026 - 23:59\",\n          \"15/07/2026 - 23:59\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Link to Content\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 200,\n        \"samples\": [\n          \"https://eu-careers.europa.eu/en/job-opportunities/policy-officer/easa-19895\",\n          \"https://eu-careers.europa.eu/en/job-opportunities/programme-manager/frontex-19815\",\n          \"https://eu-careers.europa.eu/en/job-opportunities/data-analyst/frontex-19830\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load the real data\n",
        "real_df = pd.read_csv(\"CSV-JOBS.csv\")\n",
        "print(f\"Real jobs loaded: {len(real_df)}\")\n",
        "\n",
        "# Combine real + synthetic into one dataset\n",
        "jobs_df = pd.concat([real_df, synthetic_df], ignore_index=True)\n",
        "\n",
        "print(f\"Total combined dataset: {len(jobs_df)} rows\")\n",
        "print(\"Columns:\", jobs_df.columns.tolist())\n",
        "\n",
        "display(jobs_df.head())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 505
        },
        "id": "jyIW1KljDlXP",
        "outputId": "394d684c-df7c-45d8-bb3b-83d5b83f0c0e"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Real jobs loaded: 103\n",
            "Total combined dataset: 303 rows\n",
            "Columns: ['ID', 'title', 'Domain(s)', 'Grade', 'Type of contract', 'Institution(s)    ', 'Location(s)', 'Deadline ', 'Link to Content']\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "      ID                                              title  \\\n",
              "0  19702                                  Project Assistant   \n",
              "1  19703                                     Policy Officer   \n",
              "2  19664  Senior Military Advisor to the Executive Director   \n",
              "3  19665                                  Structures Expert   \n",
              "4  19666  Certification Expert - Hydromechanical and Fli...   \n",
              "\n",
              "                  Domain(s)   Grade Type of contract  \\\n",
              "0  Justice and human rights  FG III   Contract staff   \n",
              "1  Justice and human rights   FG IV   Contract staff   \n",
              "2                 Transport   AD 10  Temporary staff   \n",
              "3                 Transport    AD 7  Temporary staff   \n",
              "4                 Transport    AD 7  Temporary staff   \n",
              "\n",
              "                                  Institution(s)            Location(s)  \\\n",
              "0  (FRA) European Union Agency for Fundamental Ri...   Vienna (Austria)   \n",
              "1  (FRA) European Union Agency for Fundamental Ri...   Vienna (Austria)   \n",
              "2       (EASA) European Union Aviation Safety Agency  Cologne (Germany)   \n",
              "3       (EASA) European Union Aviation Safety Agency  Cologne (Germany)   \n",
              "4       (EASA) European Union Aviation Safety Agency  Cologne (Germany)   \n",
              "\n",
              "            Deadline                                     Link to Content  \n",
              "0  30/04/2026 - 13:00  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "1  30/04/2026 - 13:00  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "2  30/04/2026 - 23:59  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "3  30/04/2026 - 23:59  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "4  30/04/2026 - 23:59  https://eu-careers.europa.eu/en/job-opportunit...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2d5843f3-5a2e-43a9-9098-8c9ee1215ecc\" 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>title</th>\n",
              "      <th>Domain(s)</th>\n",
              "      <th>Grade</th>\n",
              "      <th>Type of contract</th>\n",
              "      <th>Institution(s)</th>\n",
              "      <th>Location(s)</th>\n",
              "      <th>Deadline</th>\n",
              "      <th>Link to Content</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>19702</td>\n",
              "      <td>Project Assistant</td>\n",
              "      <td>Justice and human rights</td>\n",
              "      <td>FG III</td>\n",
              "      <td>Contract staff</td>\n",
              "      <td>(FRA) European Union Agency for Fundamental Ri...</td>\n",
              "      <td>Vienna (Austria)</td>\n",
              "      <td>30/04/2026 - 13:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>19703</td>\n",
              "      <td>Policy Officer</td>\n",
              "      <td>Justice and human rights</td>\n",
              "      <td>FG IV</td>\n",
              "      <td>Contract staff</td>\n",
              "      <td>(FRA) European Union Agency for Fundamental Ri...</td>\n",
              "      <td>Vienna (Austria)</td>\n",
              "      <td>30/04/2026 - 13:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>19664</td>\n",
              "      <td>Senior Military Advisor to the Executive Director</td>\n",
              "      <td>Transport</td>\n",
              "      <td>AD 10</td>\n",
              "      <td>Temporary staff</td>\n",
              "      <td>(EASA) European Union Aviation Safety Agency</td>\n",
              "      <td>Cologne (Germany)</td>\n",
              "      <td>30/04/2026 - 23:59</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>19665</td>\n",
              "      <td>Structures Expert</td>\n",
              "      <td>Transport</td>\n",
              "      <td>AD 7</td>\n",
              "      <td>Temporary staff</td>\n",
              "      <td>(EASA) European Union Aviation Safety Agency</td>\n",
              "      <td>Cologne (Germany)</td>\n",
              "      <td>30/04/2026 - 23:59</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>19666</td>\n",
              "      <td>Certification Expert - Hydromechanical and Fli...</td>\n",
              "      <td>Transport</td>\n",
              "      <td>AD 7</td>\n",
              "      <td>Temporary staff</td>\n",
              "      <td>(EASA) European Union Aviation Safety Agency</td>\n",
              "      <td>Cologne (Germany)</td>\n",
              "      <td>30/04/2026 - 23:59</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</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-2d5843f3-5a2e-43a9-9098-8c9ee1215ecc')\"\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-2d5843f3-5a2e-43a9-9098-8c9ee1215ecc 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-2d5843f3-5a2e-43a9-9098-8c9ee1215ecc');\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\": \"display(jobs_df\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"ID\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 20,\n        \"min\": 19664,\n        \"max\": 19703,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          19703,\n          19666,\n          19664\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"title\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"Policy Officer\",\n          \"Certification Expert - Hydromechanical and Flight Control Systems\",\n          \"Senior Military Advisor to the Executive Director\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Domain(s)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Transport\",\n          \"Justice and human rights\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Grade\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"FG IV\",\n          \"AD 7\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Type of contract\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Temporary staff\",\n          \"Contract staff\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Institution(s)    \",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"(EASA) European Union Aviation Safety Agency\",\n          \"(FRA) European Union Agency for Fundamental Rights\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Location(s)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Cologne (Germany)\",\n          \"Vienna (Austria)\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Deadline \",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"30/04/2026 - 23:59\",\n          \"30/04/2026 - 13:00\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Link to Content\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"https://eu-careers.europa.eu/en/job-opportunities/policy-officer/fra-ca-polof-fgiv-2026\",\n          \"https://eu-careers.europa.eu/en/job-opportunities/certification-expert-hydromechanical-and-flight-control-systems/easa-ad-2026-997\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "jobs_clean = jobs_df.copy()\n",
        "\n",
        "jobs_clean.columns = jobs_clean.columns.str.strip()\n",
        "jobs_clean[\"Deadline\"] = pd.to_datetime(jobs_clean[\"Deadline\"], errors=\"coerce\")\n",
        "\n",
        "jobs_clean = jobs_clean.dropna(subset=[\"title\", \"Domain(s)\", \"Type of contract\", \"Deadline\"])\n",
        "\n",
        "print(\"Cleaned shape:\", jobs_clean.shape)\n",
        "display(jobs_clean.head())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 505
        },
        "id": "UgMfW50wDm1g",
        "outputId": "a288151f-da02-4fdf-d3ae-8ecf1c6be725"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cleaned shape: (299, 9)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/tmp/ipykernel_12223/2873856445.py:4: UserWarning: Parsing dates in %d/%m/%Y - %H:%M format when dayfirst=False (the default) was specified. Pass `dayfirst=True` or specify a format to silence this warning.\n",
            "  jobs_clean[\"Deadline\"] = pd.to_datetime(jobs_clean[\"Deadline\"], errors=\"coerce\")\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "      ID                                              title  \\\n",
              "0  19702                                  Project Assistant   \n",
              "1  19703                                     Policy Officer   \n",
              "2  19664  Senior Military Advisor to the Executive Director   \n",
              "3  19665                                  Structures Expert   \n",
              "4  19666  Certification Expert - Hydromechanical and Fli...   \n",
              "\n",
              "                  Domain(s)   Grade Type of contract  \\\n",
              "0  Justice and human rights  FG III   Contract staff   \n",
              "1  Justice and human rights   FG IV   Contract staff   \n",
              "2                 Transport   AD 10  Temporary staff   \n",
              "3                 Transport    AD 7  Temporary staff   \n",
              "4                 Transport    AD 7  Temporary staff   \n",
              "\n",
              "                                      Institution(s)        Location(s)  \\\n",
              "0  (FRA) European Union Agency for Fundamental Ri...   Vienna (Austria)   \n",
              "1  (FRA) European Union Agency for Fundamental Ri...   Vienna (Austria)   \n",
              "2       (EASA) European Union Aviation Safety Agency  Cologne (Germany)   \n",
              "3       (EASA) European Union Aviation Safety Agency  Cologne (Germany)   \n",
              "4       (EASA) European Union Aviation Safety Agency  Cologne (Germany)   \n",
              "\n",
              "             Deadline                                    Link to Content  \n",
              "0 2026-04-30 13:00:00  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "1 2026-04-30 13:00:00  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "2 2026-04-30 23:59:00  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "3 2026-04-30 23:59:00  https://eu-careers.europa.eu/en/job-opportunit...  \n",
              "4 2026-04-30 23:59:00  https://eu-careers.europa.eu/en/job-opportunit...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-fbd49b0f-f509-4c22-8bc4-f5737d95fee2\" 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>title</th>\n",
              "      <th>Domain(s)</th>\n",
              "      <th>Grade</th>\n",
              "      <th>Type of contract</th>\n",
              "      <th>Institution(s)</th>\n",
              "      <th>Location(s)</th>\n",
              "      <th>Deadline</th>\n",
              "      <th>Link to Content</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>19702</td>\n",
              "      <td>Project Assistant</td>\n",
              "      <td>Justice and human rights</td>\n",
              "      <td>FG III</td>\n",
              "      <td>Contract staff</td>\n",
              "      <td>(FRA) European Union Agency for Fundamental Ri...</td>\n",
              "      <td>Vienna (Austria)</td>\n",
              "      <td>2026-04-30 13:00:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>19703</td>\n",
              "      <td>Policy Officer</td>\n",
              "      <td>Justice and human rights</td>\n",
              "      <td>FG IV</td>\n",
              "      <td>Contract staff</td>\n",
              "      <td>(FRA) European Union Agency for Fundamental Ri...</td>\n",
              "      <td>Vienna (Austria)</td>\n",
              "      <td>2026-04-30 13:00:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>19664</td>\n",
              "      <td>Senior Military Advisor to the Executive Director</td>\n",
              "      <td>Transport</td>\n",
              "      <td>AD 10</td>\n",
              "      <td>Temporary staff</td>\n",
              "      <td>(EASA) European Union Aviation Safety Agency</td>\n",
              "      <td>Cologne (Germany)</td>\n",
              "      <td>2026-04-30 23:59:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>19665</td>\n",
              "      <td>Structures Expert</td>\n",
              "      <td>Transport</td>\n",
              "      <td>AD 7</td>\n",
              "      <td>Temporary staff</td>\n",
              "      <td>(EASA) European Union Aviation Safety Agency</td>\n",
              "      <td>Cologne (Germany)</td>\n",
              "      <td>2026-04-30 23:59:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>19666</td>\n",
              "      <td>Certification Expert - Hydromechanical and Fli...</td>\n",
              "      <td>Transport</td>\n",
              "      <td>AD 7</td>\n",
              "      <td>Temporary staff</td>\n",
              "      <td>(EASA) European Union Aviation Safety Agency</td>\n",
              "      <td>Cologne (Germany)</td>\n",
              "      <td>2026-04-30 23:59:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</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-fbd49b0f-f509-4c22-8bc4-f5737d95fee2')\"\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-fbd49b0f-f509-4c22-8bc4-f5737d95fee2 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-fbd49b0f-f509-4c22-8bc4-f5737d95fee2');\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\": \"display(jobs_clean\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"ID\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 20,\n        \"min\": 19664,\n        \"max\": 19703,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          19703,\n          19666,\n          19664\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"title\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"Policy Officer\",\n          \"Certification Expert - Hydromechanical and Flight Control Systems\",\n          \"Senior Military Advisor to the Executive Director\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Domain(s)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Transport\",\n          \"Justice and human rights\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Grade\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"FG IV\",\n          \"AD 7\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Type of contract\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Temporary staff\",\n          \"Contract staff\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Institution(s)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"(EASA) European Union Aviation Safety Agency\",\n          \"(FRA) European Union Agency for Fundamental Rights\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Location(s)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Cologne (Germany)\",\n          \"Vienna (Austria)\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Deadline\",\n      \"properties\": {\n        \"dtype\": \"date\",\n        \"min\": \"2026-04-30 13:00:00\",\n        \"max\": \"2026-04-30 23:59:00\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"2026-04-30 23:59:00\",\n          \"2026-04-30 13:00:00\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Link to Content\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"https://eu-careers.europa.eu/en/job-opportunities/policy-officer/fra-ca-polof-fgiv-2026\",\n          \"https://eu-careers.europa.eu/en/job-opportunities/certification-expert-hydromechanical-and-flight-control-systems/easa-ad-2026-997\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"Unique domains:\", jobs_clean[\"Domain(s)\"].nunique())\n",
        "print(\"Unique contract types:\", jobs_clean[\"Type of contract\"].nunique())\n",
        "print(\"Unique institutions:\", jobs_clean[\"Institution(s)\"].nunique())\n",
        "print(\"Unique locations:\", jobs_clean[\"Location(s)\"].nunique())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wN8LLnbqDo5H",
        "outputId": "157bd26d-0dee-44ed-9b65-bbdf4547d43a"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Unique domains: 44\n",
            "Unique contract types: 5\n",
            "Unique institutions: 40\n",
            "Unique locations: 33\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "comments_by_demand = {\n",
        "    \"high\": [\n",
        "        \"This role appears to be in strong demand across institutions.\",\n",
        "        \"The labor market for this profile looks highly competitive.\",\n",
        "        \"This posting suggests strong employment opportunities.\",\n",
        "        \"Demand for this skill set appears to be growing.\",\n",
        "        \"This category seems to attract significant hiring activity.\"\n",
        "    ],\n",
        "    \"stable\": [\n",
        "        \"This role appears to have steady demand.\",\n",
        "        \"The market for this profile looks relatively balanced.\",\n",
        "        \"This posting suggests moderate but stable opportunities.\",\n",
        "        \"Demand appears consistent across institutions.\",\n",
        "        \"This category seems to have regular hiring activity.\"\n",
        "    ],\n",
        "    \"low\": [\n",
        "        \"This role appears to have weaker demand.\",\n",
        "        \"The market for this profile looks more limited.\",\n",
        "        \"This posting suggests fewer employment opportunities.\",\n",
        "        \"Demand for this skill set appears relatively low.\",\n",
        "        \"This category seems to have less hiring activity.\"\n",
        "    ]\n",
        "}"
      ],
      "metadata": {
        "id": "HHqueIqZDqVe"
      },
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "jobs_enriched = jobs_clean.copy()\n",
        "\n",
        "today = pd.Timestamp.today().normalize()\n",
        "jobs_enriched[\"days_to_deadline\"] = (jobs_enriched[\"Deadline\"] - today).dt.days\n",
        "\n",
        "def urgency_from_days(days):\n",
        "    if pd.isna(days):\n",
        "        return random.randint(3, 6)\n",
        "    if days <= 7:\n",
        "        return random.randint(8, 10)\n",
        "    elif days <= 21:\n",
        "        return random.randint(5, 8)\n",
        "    else:\n",
        "        return random.randint(2, 6)\n",
        "\n",
        "jobs_enriched[\"urgency_score\"] = jobs_enriched[\"days_to_deadline\"].apply(urgency_from_days)\n",
        "\n",
        "def demand_from_urgency(score):\n",
        "    if score >= 8:\n",
        "        return random.randint(7, 10)\n",
        "    elif score >= 5:\n",
        "        return random.randint(4, 7)\n",
        "    else:\n",
        "        return random.randint(2, 5)\n",
        "\n",
        "jobs_enriched[\"job_demand_score\"] = jobs_enriched[\"urgency_score\"].apply(demand_from_urgency)\n",
        "\n",
        "def demand_label(score):\n",
        "    if score <= 3:\n",
        "        return \"low\"\n",
        "    elif score <= 6:\n",
        "        return \"stable\"\n",
        "    else:\n",
        "        return \"high\"\n",
        "\n",
        "jobs_enriched[\"demand_label\"] = jobs_enriched[\"job_demand_score\"].apply(demand_label)\n",
        "\n",
        "jobs_enriched[\"estimated_salary\"] = np.random.randint(35000, 95000, len(jobs_enriched))\n",
        "\n",
        "jobs_enriched[\"automation_risk\"] = np.random.choice(\n",
        "    [\"low\", \"medium\", \"high\"],\n",
        "    size=len(jobs_enriched),\n",
        "    p=[0.35, 0.45, 0.20]\n",
        ")\n",
        "\n",
        "jobs_enriched[\"estimated_applications\"] = np.random.randint(20, 250, len(jobs_enriched))\n",
        "\n",
        "jobs_enriched[\"job_comment\"] = jobs_enriched[\"demand_label\"].apply(\n",
        "    lambda x: random.choice(comments_by_demand[x])\n",
        ")\n",
        "\n",
        "display(jobs_enriched.head())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 521
        },
        "id": "l6PBz502Dr3a",
        "outputId": "401257dc-e7ea-4481-eb1b-15688fbcc6b2"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "      ID                                              title  \\\n",
              "0  19702                                  Project Assistant   \n",
              "1  19703                                     Policy Officer   \n",
              "2  19664  Senior Military Advisor to the Executive Director   \n",
              "3  19665                                  Structures Expert   \n",
              "4  19666  Certification Expert - Hydromechanical and Fli...   \n",
              "\n",
              "                  Domain(s)   Grade Type of contract  \\\n",
              "0  Justice and human rights  FG III   Contract staff   \n",
              "1  Justice and human rights   FG IV   Contract staff   \n",
              "2                 Transport   AD 10  Temporary staff   \n",
              "3                 Transport    AD 7  Temporary staff   \n",
              "4                 Transport    AD 7  Temporary staff   \n",
              "\n",
              "                                      Institution(s)        Location(s)  \\\n",
              "0  (FRA) European Union Agency for Fundamental Ri...   Vienna (Austria)   \n",
              "1  (FRA) European Union Agency for Fundamental Ri...   Vienna (Austria)   \n",
              "2       (EASA) European Union Aviation Safety Agency  Cologne (Germany)   \n",
              "3       (EASA) European Union Aviation Safety Agency  Cologne (Germany)   \n",
              "4       (EASA) European Union Aviation Safety Agency  Cologne (Germany)   \n",
              "\n",
              "             Deadline                                    Link to Content  \\\n",
              "0 2026-04-30 13:00:00  https://eu-careers.europa.eu/en/job-opportunit...   \n",
              "1 2026-04-30 13:00:00  https://eu-careers.europa.eu/en/job-opportunit...   \n",
              "2 2026-04-30 23:59:00  https://eu-careers.europa.eu/en/job-opportunit...   \n",
              "3 2026-04-30 23:59:00  https://eu-careers.europa.eu/en/job-opportunit...   \n",
              "4 2026-04-30 23:59:00  https://eu-careers.europa.eu/en/job-opportunit...   \n",
              "\n",
              "   days_to_deadline  urgency_score  job_demand_score demand_label  \\\n",
              "0                 0              9                 9         high   \n",
              "1                 0              8                 8         high   \n",
              "2                 0              8                10         high   \n",
              "3                 0              9                 8         high   \n",
              "4                 0              9                 7         high   \n",
              "\n",
              "   estimated_salary automation_risk  estimated_applications  \\\n",
              "0             91422             low                      75   \n",
              "1             50795          medium                     180   \n",
              "2             35860          medium                     185   \n",
              "3             73158             low                     136   \n",
              "4             89343          medium                     153   \n",
              "\n",
              "                                         job_comment  \n",
              "0  The labor market for this profile looks highly...  \n",
              "1   Demand for this skill set appears to be growing.  \n",
              "2  This category seems to attract significant hir...  \n",
              "3  This posting suggests strong employment opport...  \n",
              "4  This category seems to attract significant hir...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-11976e12-8290-4b3f-bb3c-7b01b06b0d36\" 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>title</th>\n",
              "      <th>Domain(s)</th>\n",
              "      <th>Grade</th>\n",
              "      <th>Type of contract</th>\n",
              "      <th>Institution(s)</th>\n",
              "      <th>Location(s)</th>\n",
              "      <th>Deadline</th>\n",
              "      <th>Link to Content</th>\n",
              "      <th>days_to_deadline</th>\n",
              "      <th>urgency_score</th>\n",
              "      <th>job_demand_score</th>\n",
              "      <th>demand_label</th>\n",
              "      <th>estimated_salary</th>\n",
              "      <th>automation_risk</th>\n",
              "      <th>estimated_applications</th>\n",
              "      <th>job_comment</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>19702</td>\n",
              "      <td>Project Assistant</td>\n",
              "      <td>Justice and human rights</td>\n",
              "      <td>FG III</td>\n",
              "      <td>Contract staff</td>\n",
              "      <td>(FRA) European Union Agency for Fundamental Ri...</td>\n",
              "      <td>Vienna (Austria)</td>\n",
              "      <td>2026-04-30 13:00:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "      <td>0</td>\n",
              "      <td>9</td>\n",
              "      <td>9</td>\n",
              "      <td>high</td>\n",
              "      <td>91422</td>\n",
              "      <td>low</td>\n",
              "      <td>75</td>\n",
              "      <td>The labor market for this profile looks highly...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>19703</td>\n",
              "      <td>Policy Officer</td>\n",
              "      <td>Justice and human rights</td>\n",
              "      <td>FG IV</td>\n",
              "      <td>Contract staff</td>\n",
              "      <td>(FRA) European Union Agency for Fundamental Ri...</td>\n",
              "      <td>Vienna (Austria)</td>\n",
              "      <td>2026-04-30 13:00:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>8</td>\n",
              "      <td>high</td>\n",
              "      <td>50795</td>\n",
              "      <td>medium</td>\n",
              "      <td>180</td>\n",
              "      <td>Demand for this skill set appears to be growing.</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>19664</td>\n",
              "      <td>Senior Military Advisor to the Executive Director</td>\n",
              "      <td>Transport</td>\n",
              "      <td>AD 10</td>\n",
              "      <td>Temporary staff</td>\n",
              "      <td>(EASA) European Union Aviation Safety Agency</td>\n",
              "      <td>Cologne (Germany)</td>\n",
              "      <td>2026-04-30 23:59:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>10</td>\n",
              "      <td>high</td>\n",
              "      <td>35860</td>\n",
              "      <td>medium</td>\n",
              "      <td>185</td>\n",
              "      <td>This category seems to attract significant hir...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>19665</td>\n",
              "      <td>Structures Expert</td>\n",
              "      <td>Transport</td>\n",
              "      <td>AD 7</td>\n",
              "      <td>Temporary staff</td>\n",
              "      <td>(EASA) European Union Aviation Safety Agency</td>\n",
              "      <td>Cologne (Germany)</td>\n",
              "      <td>2026-04-30 23:59:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "      <td>0</td>\n",
              "      <td>9</td>\n",
              "      <td>8</td>\n",
              "      <td>high</td>\n",
              "      <td>73158</td>\n",
              "      <td>low</td>\n",
              "      <td>136</td>\n",
              "      <td>This posting suggests strong employment opport...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>19666</td>\n",
              "      <td>Certification Expert - Hydromechanical and Fli...</td>\n",
              "      <td>Transport</td>\n",
              "      <td>AD 7</td>\n",
              "      <td>Temporary staff</td>\n",
              "      <td>(EASA) European Union Aviation Safety Agency</td>\n",
              "      <td>Cologne (Germany)</td>\n",
              "      <td>2026-04-30 23:59:00</td>\n",
              "      <td>https://eu-careers.europa.eu/en/job-opportunit...</td>\n",
              "      <td>0</td>\n",
              "      <td>9</td>\n",
              "      <td>7</td>\n",
              "      <td>high</td>\n",
              "      <td>89343</td>\n",
              "      <td>medium</td>\n",
              "      <td>153</td>\n",
              "      <td>This category seems to attract significant hir...</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-11976e12-8290-4b3f-bb3c-7b01b06b0d36')\"\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-11976e12-8290-4b3f-bb3c-7b01b06b0d36 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-11976e12-8290-4b3f-bb3c-7b01b06b0d36');\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\": \"display(jobs_enriched\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"ID\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 20,\n        \"min\": 19664,\n        \"max\": 19703,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          19703,\n          19666,\n          19664\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"title\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"Policy Officer\",\n          \"Certification Expert - Hydromechanical and Flight Control Systems\",\n          \"Senior Military Advisor to the Executive Director\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Domain(s)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Transport\",\n          \"Justice and human rights\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Grade\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"FG IV\",\n          \"AD 7\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Type of contract\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Temporary staff\",\n          \"Contract staff\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Institution(s)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"(EASA) European Union Aviation Safety Agency\",\n          \"(FRA) European Union Agency for Fundamental Rights\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Location(s)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Cologne (Germany)\",\n          \"Vienna (Austria)\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Deadline\",\n      \"properties\": {\n        \"dtype\": \"date\",\n        \"min\": \"2026-04-30 13:00:00\",\n        \"max\": \"2026-04-30 23:59:00\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"2026-04-30 23:59:00\",\n          \"2026-04-30 13:00:00\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Link to Content\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"https://eu-careers.europa.eu/en/job-opportunities/policy-officer/fra-ca-polof-fgiv-2026\",\n          \"https://eu-careers.europa.eu/en/job-opportunities/certification-expert-hydromechanical-and-flight-control-systems/easa-ad-2026-997\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"days_to_deadline\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 0,\n        \"max\": 0,\n        \"num_unique_values\": 1,\n        \"samples\": [\n          0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"urgency_score\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 8,\n        \"max\": 9,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          8\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"job_demand_score\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1,\n        \"min\": 7,\n        \"max\": 10,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          8\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"demand_label\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"high\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"estimated_salary\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 24288,\n        \"min\": 35860,\n        \"max\": 91422,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          50795\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"automation_risk\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"medium\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"estimated_applications\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 44,\n        \"min\": 75,\n        \"max\": 185,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          180\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"job_comment\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"Demand for this skill set appears to be growing.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "jobs_enriched.to_csv(\"jobs_enriched.csv\", index=False)\n",
        "print(\"Saved: jobs_enriched.csv\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "s00SIZBsDtd-",
        "outputId": "e87d8cc4-4a9c-4263-ddd1-9ddc8d02cbc9"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saved: jobs_enriched.csv\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "jobs_enriched[\"deadline_month\"] = jobs_enriched[\"Deadline\"].dt.to_period(\"M\").astype(str)\n",
        "\n",
        "monthly_openings = (\n",
        "    jobs_enriched\n",
        "    .groupby([\"deadline_month\", \"Domain(s)\"], as_index=False)\n",
        "    .agg(\n",
        "        postings=(\"title\", \"count\"),\n",
        "        avg_salary=(\"estimated_salary\", \"mean\"),\n",
        "        avg_demand_score=(\"job_demand_score\", \"mean\"),\n",
        "        avg_applications=(\"estimated_applications\", \"mean\")\n",
        "    )\n",
        ")\n",
        "\n",
        "display(monthly_openings.head())\n",
        "print(\"Shape:\", monthly_openings.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 241
        },
        "id": "3NtBDh8ODvLW",
        "outputId": "aad6702c-076b-47cd-b398-1cf215d10268"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  deadline_month                          Domain(s)  postings    avg_salary  \\\n",
              "0        2026-04                    Data protection         1  79732.000000   \n",
              "1        2026-04  Economics, Finance and Statistics         1  46284.000000   \n",
              "2        2026-04                    Human Resources         1  89886.000000   \n",
              "3        2026-04           Justice and human rights         2  71108.500000   \n",
              "4        2026-04                          Transport         3  66120.333333   \n",
              "\n",
              "   avg_demand_score  avg_applications  \n",
              "0          9.000000              77.0  \n",
              "1          7.000000              63.0  \n",
              "2          7.000000             192.0  \n",
              "3          8.500000             127.5  \n",
              "4          8.333333             158.0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-d89ad582-a695-47da-9deb-fe333bb72f92\" 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>deadline_month</th>\n",
              "      <th>Domain(s)</th>\n",
              "      <th>postings</th>\n",
              "      <th>avg_salary</th>\n",
              "      <th>avg_demand_score</th>\n",
              "      <th>avg_applications</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2026-04</td>\n",
              "      <td>Data protection</td>\n",
              "      <td>1</td>\n",
              "      <td>79732.000000</td>\n",
              "      <td>9.000000</td>\n",
              "      <td>77.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2026-04</td>\n",
              "      <td>Economics, Finance and Statistics</td>\n",
              "      <td>1</td>\n",
              "      <td>46284.000000</td>\n",
              "      <td>7.000000</td>\n",
              "      <td>63.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2026-04</td>\n",
              "      <td>Human Resources</td>\n",
              "      <td>1</td>\n",
              "      <td>89886.000000</td>\n",
              "      <td>7.000000</td>\n",
              "      <td>192.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2026-04</td>\n",
              "      <td>Justice and human rights</td>\n",
              "      <td>2</td>\n",
              "      <td>71108.500000</td>\n",
              "      <td>8.500000</td>\n",
              "      <td>127.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2026-04</td>\n",
              "      <td>Transport</td>\n",
              "      <td>3</td>\n",
              "      <td>66120.333333</td>\n",
              "      <td>8.333333</td>\n",
              "      <td>158.0</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-d89ad582-a695-47da-9deb-fe333bb72f92')\"\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-d89ad582-a695-47da-9deb-fe333bb72f92 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-d89ad582-a695-47da-9deb-fe333bb72f92');\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\": \"print(\\\"Shape:\\\", monthly_openings\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"deadline_month\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"2026-04\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Domain(s)\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"Economics, Finance and Statistics\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"postings\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 1,\n        \"max\": 3,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"avg_salary\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 16331.975065972749,\n        \"min\": 46284.0,\n        \"max\": 89886.0,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          46284.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"avg_demand_score\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.9159087776022724,\n        \"min\": 7.0,\n        \"max\": 9.0,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          7.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"avg_applications\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 54.13178363955875,\n        \"min\": 63.0,\n        \"max\": 192.0,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          63.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Shape: (138, 6)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "monthly_openings.to_csv(\"jobs_monthly_openings.csv\", index=False)\n",
        "print(\"Saved: jobs_monthly_openings.csv\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_WpGUgWSDxDr",
        "outputId": "569af645-b25b-40f8-f70b-522e5c8be3de"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saved: jobs_monthly_openings.csv\n"
          ]
        }
      ]
    }
  ]
}