🎬 AI-Powered Movie Recommendation System: Discover Movies You’ll Love

Finding a good movie can often feel overwhelming with thousands of choices available online. To solve this, Mirza Yasir Abdullah Baig created the Movie Recommendation System, an AI-powered web application that recommends movies similar to the ones you already enjoy. Built with Streamlit, Scikit-learn, and OMDb API, this project showcases the power of machine learning and content-based recommendation systems.

Also Read:

Image Classification Model By Mirza Yasir Abdullah Baig

Loan Status Prediction Model By Mirza Yasir Abdullah Baig

Movie Recommendation Model By Mirza Yasir Abdullah Baig

Sentiment Analysis Model By Mirza Yasir Abdullah Baig

StockMarket Trend Prediction Model By Mirza Yasir Abdullah Baig

🚀 Overview

The Movie Recommendation System is an intelligent web app designed to analyze movie metadata and suggest the top 5 similar movies based on your chosen title. Using a cosine similarity-based content filtering model, it compares features like genres, keywords, and overviews to identify movies that share similar characteristics.

This application combines simplicity and performance — providing instant, data-driven movie suggestions within a sleek, user-friendly Streamlit interface.

🎥 How It Works

1. The user selects a movie from a dropdown list in the web app.
2. The system retrieves the movie’s metadata from a preprocessed dataset (movies.csv).
3. Using the TF-IDF Vectorizer and Cosine Similarity, the model calculates how closely related other movies are.
4. The OMDb API is then called to fetch posters and details of the top 5 recommendations.
5. Results are displayed dynamically in the app — complete with titles and movie posters.

📊 Dataset & Model

• Dataset: movies.csv (contains movie titles, genres, overviews, and metadata)
• Model: Precomputed cosine similarity matrix stored in tfidf_matrix.pkl
• APIs: OMDb API for fetching real-time movie posters and information
• Scripts: preprocess.py, recommend.py, model.py, and main.py orchestrate data cleaning, model building, and app functionality

⚙️ Tech Stack

• Python 3.9+ – Core programming language
• Streamlit – Frontend web app framework
• Pandas & NumPy – Data handling and numerical computation
• Scikit-learn – TF-IDF and cosine similarity calculations
• Requests – API requests for fetching movie posters from OMDb

🌟 Key Features

• 🎞️ Top 5 Personalized Movie Suggestions based on similarity
• 🖼️ Dynamic Poster Fetching from the OMDb API
• ⚡ Fast & Accurate Recommendations using cosine similarity
• 💻 User-Friendly Streamlit Interface
• 👤 Developer Sidebar with portfolio and social links

🧠 Behind the Model

This system uses content-based filtering, a classic approach to recommendations. Instead of relying on user ratings or collaborative behavior, it focuses on the content itself — matching movies based on textual and categorical similarities.

For instance, if you liked Inception, the model might recommend Interstellar or The Matrix based on similar genres, keywords, and descriptions. This makes the experience highly personalized and interpretable.

📺 Live Demo

You can explore the app yourself here:
🔗 Movie Recommendation App on Streamlit

Watch how it works: 🎥 movie-recommendation.webm

👨‍💻 About the Developer

Mirza Yasir Abdullah Baig is an AI and Machine Learning Engineer passionate about creating intelligent applications that merge data science with real-world usability.
You can connect with him through:
🌐 Kaggle
💼 LinkedIn
💻 GitHub

❤️ Acknowledgements

• OMDb API
• Streamlit Documentation
• Scikit-learn

⚠️ Disclaimer

This project is purely for educational and research purposes, built to demonstrate the working of content-based movie recommendation systems using AI and Python.

Final Thoughts:

The Movie Recommendation System is a perfect blend of AI, creativity, and user experience — showing how machine learning can make everyday entertainment smarter. Whether you’re a developer exploring recommendation algorithms or a movie lover looking for your next favorite film, this project is worth checking out.