Flix is a movie recommender app which utilizes the MovieLens database to provide personalized movie recommendations based on user preferences. The app employs a collaborative filtering technique, leveraging a correlation matrix derived from user ratings. It selects the top-rated movies and those with a substantial number of reviews to build a correlation matrix, which forms the backbone of the recommendation engine.
The recommendation engine's machine learning model is implemented using Python, leveraging libraries such as numPy, MathPlotLib, Seaborn, and SciKit-Learn. Python is used for the data analysis tasks, including building correlation matrices and exporting them as pkls for further processing.
- Recommendation Engine: Generates movie recommendations based on user preferences and ratings.
- Live Search: Allows users to search for movies in real-time using a live search feature.
- Data Visualization: Visualizes movie data using various graphs and charts to provide insights into movie ratings, genres, and other relevant information.
- Database Integration: Utilizes MySQL for storing movie names and facilitating live querying and seamless integration with the recommendation engine.
- Frontend: React, Tailwind CSS.
- Backend: Vite, Node, FastAPI.
- Database: MySQL.
- Data Analysis and Machine Learning Model: Python, numPy, MathPlotLib, Seaborn, SciKit-Learn.
- Data Collection: The app retrieves movie data from the MovieLens database, including user ratings and movie information.
- Correlation Matrix Generation: It constructs a correlation matrix based on user ratings, focusing on highly-rated movies and those with a significant number of reviews.
- Recommendation Generation: Using the correlation matrix, the app generates movie recommendations tailored to each user's preferences.
- Live Search: Users can search for movies in real-time using the live search feature, which fetches results dynamically as the user types.
- Data Visualization: The app visualizes movie data using various graphs and charts, providing users with insights into movie ratings, genres, and other relevant metrics.
This project is licensed under the MIT License - see the LICENSE file for details.