0

AI-Assisted Football Referee

https://towardsdatascience.com/ai-assisted-football-referee/(towardsdatascience.com)
An AI-assisted football referee system is presented to answer questions about fouls by referencing the official IFAB Laws of the Game. The project uses a Retrieval-Augmented Generation (RAG) architecture to ensure answers are based on the source document, not the model's memory, thus avoiding hallucination. The technical implementation involves extracting text with PyMuPDF, creating embeddings with Hugging Face, storing them in ChromaDB, and generating final answers with the Gemini LLM. A key feature is the use of a parent-child chunking method to maintain the context of the rules during data processing.
0 points•by ogg•1 hour ago

Comments (0)

No comments yet. Be the first to comment!

Have an account? Log in to join the discussion.