Skip to content

README.md

Latest commit

 

History

History
123 lines (86 loc) · 2.97 KB

README.md

File metadata and controls

123 lines (86 loc) · 2.97 KB

🎬 FlixRecommender

Netflix Clustering-Based Recommender System

An interactive Streamlit app that clusters and recommends Netflix shows using unsupervised machine learning. It applies K-Means clustering on genres and duration, supports search with fuzzy matching, and integrates poster images and trailers via the TMDB API.

📸 Preview

App Screenshot

🚀 Features

  • 🔍 Smart Search with fuzzy matching using rapidfuzz
  • 🎯 K-Means Clustering based on show genres and duration
  • 🤖 Intelligent Recommendations of similar shows from the same cluster
  • 📸 Live Show Posters from TMDB API
  • 🎞️ Trailers & Full Descriptions fetched dynamically
  • 🧪 PCA Scatter Plot to visualize cluster distribution
  • 🎛️ Advanced Filters: Filter by rating, country, or release year

🚀 Live Deploed Model

👉 flixrecommender.streamlit.app

📁 Project Structure

trahulsingh-flixrecommender/
├── app.py                  # 🎯 Main Streamlit app
├── requirements.txt        # 📦 All required Python dependencies
├── README.md               # 📘 This file
├── LICENSE                 # 📄 MIT License
├── .env                    # 🔐 API keys (e.g., TMDB_API_KEY)
├── maincode.ipynb         # 📓 (Optional) Jupyter version
└── data/
    └── netflix_titles.csv  # 📄 Dataset from Kaggle

📦 Installation

  1. Clone the repository
git clone https://github.com/TRahulsingh/FlixRecommender.git
cd FlixRecommender
  1. Install dependencies
pip install -r requirements.txt
  1. Configure TMDB API key
    Create a .env file in the root directory:
TMDB_API_KEY=your_tmdb_api_key_here

Get your API key from: TMDB API

  1. Run the Streamlit app
streamlit run app.py

📈 Dataset

This app uses the public Netflix Movies and TV Shows dataset from Kaggle:
📥 https://www.kaggle.com/datasets/shivamb/netflix-shows

The dataset includes:

  • 🎬 Title
  • 📅 Release year
  • 🕒 Duration
  • 🌍 Country
  • 🔞 Rating
  • 🎭 Listed genres
  • 📝 Description

🛠 Built With

  • Python 🐍
  • Streamlit ⚡
  • Scikit-learn – KMeans, PCA
  • RapidFuzz – for fuzzy title matching
  • Pandas & NumPy – data handling
  • Matplotlib & Seaborn – data visualization
  • TMDB API – for posters, metadata, and trailers
  • dotenv – to securely manage API keys

🧠 Future Improvements

  • 🔒 User login and personal watchlists
  • 📊 Add content-based or collaborative filtering

🧑‍💻 Author

Made with ❤️ by TRahulsingh
GitHub: @TRahulsingh

📄 License

This project is licensed under the MIT License. See the LICENSE file for details.
Feel free to fork, modify, and share!