https://github.com/shivamratti13/movie_recommendation_sys
https://github.com/shivamratti13/movie_recommendation_sys
jupyternotebook movies-streamlit python recommender-system
Last synced: 28 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/shivamratti13/movie_recommendation_sys
- Owner: shivamratti13
- Created: 2024-10-23T11:53:24.000Z (about 1 year ago)
- Default Branch: master
- Last Pushed: 2025-01-13T12:38:18.000Z (10 months ago)
- Last Synced: 2025-03-23T22:32:02.693Z (7 months ago)
- Topics: jupyternotebook, movies-streamlit, python, recommender-system
- Language: Jupyter Notebook
- Homepage: https://movierecommendationsys-zxtnlxpqj3rglmzsf5gmet.streamlit.app/
- Size: 8.68 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
In my movie recommendation project, I utilized the TMDB dataset from Kaggle to build a content-based recommendation system. The goal of the project was to suggest five similar movies to users based on the features of the movie they liked. Content-based filtering focuses on the attributes of items, in this case, movies, to make recommendations by comparing features such as genres, cast, crew, keywords, and plot overviews.
I started by cleaning and preprocessing the dataset, ensuring that missing or irrelevant data points were handled effectively. To compute the similarity between movies, I employed techniques such as the bag of words vectorization on the plot descriptions, which helped transform the text into numerical vectors. Using cosine similarity, I measured the closeness between movies, ensuring that the most similar ones were identified based on their feature vectors.
The recommendation engine then provided users with a list of five similar movies whenever they selected or searched for a particular movie. The model was evaluated and fine-tuned to ensure relevant and accurate recommendations, improving the overall user experience. This project demonstrated my skills in data preprocessing, feature extraction, and machine learning, highlighting my ability to implement real-world data-driven applications.
