Kaggle Jigsaw - Agile Community Rules Classification

This repository contains our solution for the Kaggle Jigsaw - Agile Community Rules Classification competition.

Competition Overview

The Jigsaw Agile Community Rules Classification competition challenges participants to build binary classifiers that predict whether a Reddit comment violates specific subreddit rules. The goal is to assist moderators in upholding community-specific norms by automatically identifying rule violations.

Key Details:

• Task: Binary classification to predict rule violation probability
• Dataset: Moderated Reddit comments with hypothetical rules
• Evaluation: Accuracy in identifying rule violations
• Challenge: Diverse norms and expectations across different subreddits

Repository Structure

├── data/                 # Competition datasets and processed data
├── notebooks/           # Jupyter notebooks for analysis and modeling
├── src/                # Source code for preprocessing and modeling
├── models/             # Trained models and artifacts
├── results/            # Performance metrics and analysis results
├── config/             # Configuration files
└── requirements.txt    # Python dependencies

Getting Started

Prerequisites

• Python 3.8+
• Git

Installation

1. Clone the repository:

   git clone <repository-url>
   cd Kaggle-Jigsaw

2. Create a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Download the dataset:

   • Download the competition dataset from Kaggle
   • Place the files in the data/ directory

Usage

1. Run exploratory data analysis:

   jupyter notebook notebooks/

2. Train models:

   python src/train.py

3. Generate predictions:

   python src/predict.py

Methodology

Our approach includes:

• Data Preprocessing: Text cleaning and normalization
• Feature Engineering: Rule-specific and comment-specific feature extraction
• Model Development: Experimentation with various ML approaches
• Evaluation: Comprehensive performance assessment and validation

Competition Performance

[Performance metrics and rankings will be updated here as we progress]

Contributing

This is a competition repository. Please refer to the competition rules and guidelines.

Acknowledgments

• Kaggle and Jigsaw for organizing this competition
• The Reddit community for providing the dataset
• Previous Jigsaw competition winners for inspiration

License

This project is licensed under the MIT License - see the LICENSE file for details.

---

Competition Link: Kaggle Jigsaw - Agile Community Rules Classification