Graph-Based Relationships
Hybrid search has limitations in capturing relationships between entities, particularly when using traditional embeddings. By modeling connections explicitly as triples, where entities are linked through defined relationships, one can enhance the retrieval process. This approach allows for a richer context in language model generation, effectively combining dense embeddings with graph traversal for improved accuracy in answers.In this clip
From this podcast
Practical AI
GraphRAG (beyond the hype)
Related Questions
How do vector embeddings work?
I'm thinking that classical or deep ML solutions have a flaw in that they cannot be extended. For example, if one builds a model and wants to introduce a new feature, typically the model has to be retrained from scratch. So, I'm considering building a Knowledge Graph using LLMs. This knowledge graph would have to include time-dependent data (for example, it should be able to retrieve the "current" President of the USA and also previous presidents if asked). I'm thinking this Knowledge Graph could be used for Retrieval Augmented Generation (RAG) to help with business goals or maybe used with more clever User Interfaces (UIs). I'm not sure how to build or populate this KG and also have a rough idea of how to use it. Can you help me with suggesting particular paths for building and populating this Knowledge Graph?