🎬 CineMatch

An AI-powered Content-Based Movie Recommendation System built using Machine Learning, Streamlit, and the TMDB API.

---

📌 Overview

CineMatch helps users discover movies similar to the one they love by analyzing movie content instead of user ratings.

Using content-based filtering and cosine similarity, the system recommends the five most similar movies from a dataset of over 4,800 movies. Each recommendation is enriched with live information from the TMDB API, including posters, ratings, genres, runtime, director, language, and movie overview.

---

🚀 Live Demo

https://harsh-cinematch.onrender.com/

📸 Preview

Home Page

Search Movie

Movie Details

Recommendations

---

✨ Features

• 🎥 Content-Based Movie Recommendation
• 🔍 Search from 4,800+ Movies
• ⚡ Fast Recommendations using Cosine Similarity
• 🖼️ Movie Posters from TMDB
• ⭐ Movie Ratings
• 🎭 Genres
• 🎬 Director Information
• 👥 Top Cast
• 🌐 Original Language
• 📅 Release Year
• ⏱ Runtime
• 📝 Movie Overview
• 🎨 Modern Streamlit Interface
• 📱 Responsive Layout

---

🧠 Recommendation Workflow

User selects a movie
        │
        ▼
Content-Based Filtering
        │
        ▼
CountVectorizer Features
        │
        ▼
Cosine Similarity
        │
        ▼
Top 5 Similar Movies
        │
        ▼
TMDB API
        │
        ▼
Movie Posters & Details

---

🛠️ Tech Stack

Category	Technology
Language	Python
Framework	Streamlit
Machine Learning	Scikit-learn
Data Processing	Pandas
API	TMDB API
Model Storage	Pickle

---

📂 Project Structure

CineMatch/
│
├── app.py
├── assets/
│   ├── styles.css
│   └── no_poster_avail.jpg
│
├── data/
│   ├── tmdb_5000_movies.csv
│   └── tmdb_5000_credits.csv
│
├── models/
│   ├── movies.pkl
│   └── similarity.pkl
│
├── notebook/
│   └── CineMatch.ipynb
│
├── screenshots/
│   ├── home.png
│   ├── search.png
│   ├── recommendations.png
│   └── movie_details.png
│
├── utils/
│   └── tmdb.py
│
├── requirements.txt
├── .gitignore
├── README.md
└── .env.example

---

🚀 Installation

Clone the repository

git clone https://github.com/harsh8767/CineMatch.git

Navigate to the project

cd CineMatch

Install dependencies

pip install -r requirements.txt

Create a .env file

TMDB_API_KEY=your_tmdb_api_key

Run the application

streamlit run app.py

---

🧠 Machine Learning Workflow

• Load TMDB Movie Dataset
• Data Cleaning & Preprocessing
• Feature Engineering
• Tags Generation
• Text Vectorization using CountVectorizer
• Cosine Similarity Calculation
• Save Model using Pickle
• Build Interactive UI with Streamlit

---

📊 Dataset

This project uses the TMDB 5000 Movie Dataset, consisting of:

• TMDB 5000 Movies Dataset
• TMDB 5000 Credits Dataset

---

⚠️ Known Limitations

• Uses only content-based filtering.
• No personalized user recommendations.
• Requires a valid TMDB API key.
• Recommendations are limited to the dataset.

---

🌟 Future Improvements

• ❤️ Favorites
• 📚 Watchlist
• 👤 User Authentication
• 🎬 TV Show Recommendations
• 🤝 Collaborative Filtering
• 🔥 Hybrid Recommendation System
• 🌙 Dark / Light Theme Toggle

---

🙏 Acknowledgements

• TMDB API
• TMDB 5000 Movie Dataset
• Streamlit
• Scikit-learn
• Pandas

---

👨‍💻 Developer

Harsh Chavan

Computer Engineering Graduate

GitHub: https://github.com/harsh8767

---

📜 License

This project is licensed under the MIT License.