MovieMate

An AI-powered web app that connects users with personalized and meaningful movie recommendations through semantic search, enhancing the entertainment discovery experience.

Inspiration

We encountered several challenges, such as short timelines, slow team formation, and unexpected external factors. However, after attending the RAG AI (Retrieval-Augmented Generation) session, we refocused on making an impact. Our goal was to enhance Netflix's movie recommendations by applying semantic search to their database. Our team consisted of two graduate students and one undergraduate.

What it does

MovieMate is an online tool that uses semantic search to provide users with insightful movie recommendations and reviews. By integrating a powerful semantic search engine trained on the Netflix/IMDB dataset, the platform enables users to discover films based on intricate relationships between genres, plotlines, and reviews. This creates a more tailored and immersive movie-watching experience. How we built it We built the front end of MovieMate using the Streamlit framework, and the backend is powered by Python, which handles data processing and search optimization. The core of the application is a semantic search engine trained on the Netflix/IMDB dataset, ensuring fast and accurate movie recommendations for users.

Challenges we faced

Throughout the development process, we encountered several obstacles: Limited time due to tight deadlines. Initial challenges in team coordination. External factors that slowed progress. Difficulty in determining the project's technical scope and framework.

Accomplishments we’re proud of

Progress on a Functional Web Application: We’re proud of the progress we’ve made in developing a web app that uses semantic search to provide personalized movie recommendations.

Successful Integration of Streamlit: Our use of Streamlit allowed us to create a user-friendly interface for movie discovery.

Tailoring Semantic Search for Movies: By refining semantic search for entertainment purposes, we aim to enhance how users discover films and related content.

Collaborative Effort: Our team, consisting of both undergraduate and graduate students, fostered an environment of creative exchange that was key to the success of this project.

What we learned

This project reinforced the importance of adaptability and teamwork. Additionally, we improved our skills in semantic search technologies, learning how to optimize them for performance and handle large, complex movie datasets.

What’s next for MovieMate?

Looking forward, we plan to: Improve the semantic search engine to provide even more accurate and relevant recommendations. Incorporate user feedback for continuous improvement. Partner with movie streaming platforms to expand our database and reach. Introduce enhanced movie discovery features and personalized user experiences to make the platform even more engaging for movie enthusiasts.

Built With

• dataset
• python
• semantic-engines-semantic
• streamlit