Tastebud Theater

Movie Recommendation

Inspiration

In today's digital age, movie recommendation systems have become an integral part of our entertainment experience. However, most existing systems focus on individual preferences, often overlooking the social aspect of movie-watching. Our project aims to bridge this gap by developing a Collaborative Movie Recommendation System that takes into account the preferences of multiple users, such as friends or family members, to suggest movies that everyone will enjoy.

What it does

The project takes natural language inputs from users (e.g., "I like Sci-Fi movies") and converts them into Prolog facts. It then uses these facts to generate movie recommendations. If the required facts are missing, the system automatically generates them and adds them to the knowledge base.

How we built it

Python - ollama 3.2 SWI prolog- backend

This project builds a movie recommendation system using Ollama 3.2, which allows running large language models (LLMs) on machines with limited resources (e.g., 8GB of RAM). The system recommends movies based on shared preferences (genres, actors) between two users (such as friends).

Challenges we ran into

Data Collection: Gathering accurate and comprehensive movie data, including titles, genres, ratings, and user preferences, can be difficult. Data Format: Converting the collected data into a format suitable for Prolog, ensuring correct syntax to avoid errors, can be tricky. Learning Prolog and it's constructs.

Accomplishments that we're proud of

1)Users don't need to learn Prolog; they can simply provide input in natural language, making the system easy and intuitive to use. 2) We learned a new way of thinking in a short time. 3) We successfully added a level of a abstraction and interactivity to the sCASP system.

What we learned

We learned how to integrate Ollama 3.2 into our project and how to use it for running language models. We also gained a deeper understanding of how Prolog works, especially for logic-based programming and handling facts and rules.

What's next for Tastebud Theater

After the hackathon, we plan to take Tastebud Theater to the next level by:

Expanding the movie database with more genres and actors to make recommendations richer and more diverse.
Improving the recommendation logic by adding personalized suggestions based on user ratings and preferences.
Allowing multi-user support, so groups of friends can easily find movies they all like, making movie night planning easier for everyone.

These steps will help us create a more dynamic, user-friendly experience, and we’re excited to continue building on it!