Graph RAG and Rules-based AI

In this video, Nick Form and Cen Xi Toh from Oxford Semantic Technologies demonstrate how to enhance the accuracy of Large Language Models (LLMs) using a Retrieval Augmented Generation (RAG) approach with RDFox, a knowledge graph with an embedded semantic reasoner. They showcase a chatbot that provides expert answers to complex questions about recipes, leveraging RDFox for reasoning and rule-based AI. The demo highlights how the chatbot can understand and respond to queries about ingredients, dietary restrictions, and cooking instructions by integrating data from a knowledge graph and using rules to ensure accurate responses. The presenters explain the interplay between LLMs and RDFox, emphasizing the benefits of combining human expertise with automated reasoning to improve the reliability and explainability of AI systems. The video also explores potential applications of this technology across various industries and invites viewers to try RDFox and attend a free reasoning workshop.