Food Waste Ontology Chatbot
Overview
This project provides an interactive chatbot interface to explore and build structured ontologies for food waste. It uses Meta-Llama-3.1-8B-Instruct with Retrieval-Augmented Generation (RAG) to answer questions based on uploaded PDFs. Users can provide feedback on the model’s answers, edit them, and store evaluations in a local SQLite database.
Acknowledgements
- National Science Foundation (NSF) funded AI institute for Intelligent Cyberinfrastructure with Computational Learning in the Environment (ICICLE) (OAC 2112606)