Movie Recommender System

A content-based movie and TV series recommendation system using natural language processing and similarity matching.

Live Demo - Click here

Features

• Recommends movies and TV series based on content similarity.
• Utilizes advanced NLP techniques through sentence transformers for better content understanding.
• Supports both Hollywood and Bollywood content.
• Processes metadata such as descriptions, genres, and director information.
• Provides an interactive web interface built with Streamlit.

Tech Stack

• Python 3.12+
• Sentence Transformers: all-MiniLM-L6-v2 for embedding generation.
• Pandas: Data manipulation and analysis.
• Streamlit: Web application framework.
• NumPy: Numerical operations.

Project Structure

.
├── data/
│   ├── processed_data/
│   │   ├── movies.csv
│   │   └── series.csv
│   └── raw_data/
│       ├── bollywood_movies.csv
│       ├── bollywood_series.csv
│       ├── hollywood_movies.csv
│       └── hollywood_series.csv
├── notebooks/
│   └── data_generator.ipynb
├── src/
    └── movie_recommender_system/
        ├── config.py
        ├── data_preprocessor.py
        ├── st_app.py
        └── text_similarities.py

Installation

1. Clone the repository:

   git clone https://github.com/your-username/movie-recommender-system.git
   cd movie-recommender-system

2. Create and activate a virtual environment using uv:

   uv create .venv
   uv activate .venv

3. Install dependencies:

   pip install -e .

Usage

1. Start the Streamlit app:

   streamlit run src/movie_recommender_system/st_app.py

2. Enter a movie or TV series title in the search box.

3. Get personalized recommendations based on content similarity.