Problem Statement 6
INFO
This problem statement is shared by SingularityNET - title sponsor for HackIndia 2025.
🏆 Work on this problem statement if you want to compete in the $25,000 Prize Pool.
Project: Mini-Recommendation System with MeTTa Knowledge Representation
Overview
This project aims to build a recommendation system that leverages the MeTTa programming language for knowledge representation and querying. By structuring movie data in MeTTa, we can capture relationships between genres, directors, user preferences, and viewing history more effectively. The system will provide reasoning for each recommendation, explaining why a movie was suggested based on structured knowledge rather than just statistical correlation
Challenge
DDevelop a mini-recommendation system that:
- Uses MeTTa for structured knowledge representation of movies, genres, user preferences, and relationships.
- Supports both content-based and collaborative filtering approaches.
- Provides explanations for recommendations using knowledge querying functions.
- Integrates Python for additional utilities such as data processing and user interaction.
- Adapts to new user preferences dynamically by updating the knowledge base.
Key Features
- MeTTa-Based Knowledge Graph: Represents movies, genres, actors, and user preferences in a structured way.
- Content & Collaborative Filtering: Suggests movies based on genre similarity or user preference patterns.
- Explainable Recommendations: Justifies each recommendation by explaining relationships in the knowledge graph.
- Python Integration: Uses Python to interface with MeTTa for efficient querying and result retrieval.
Dynamic Learning
Updates the knowledge base with user feedback and new movie data.
Notice & Requirements
Participants must adhere to the following guidelines:
Knowledge Representation in MeTTa: Structure movie data to capture meaningful relationships (e.g., genre, director, actors, user ratings).Define query functions to retrieve and infer recommendations.
MeTTa-Python Integration: Python must be used to interact with the MeTTa-based knowledge graph and process results. Use python mainly for user Interaction , such as APi …
Knowledge Querying & Relationship Extraction: Develop functions in MeTTa to extract meaningful insights for recommendations. Solutions demonstrating effective reasoning and explainability will receive additional points.
Impact
This recommendation system will enhance user trust and engagement by offering transparent and well-reasoned movie suggestions. This system explains why a movie is recommended, making it more interpretable and valuable for users.
Learning Outcomes
- Designing knowledge-based recommendation systems with MeTTa.
- Querying structured knowledge for recommendation logic.
- Implementing Python-MeTTa integration for user Interaction.
- Exploring content-based and collaborative filtering techniques.