Skip to main content

KAG

KAG is a logical form-guided reasoning and retrieval framework based on OpenSPG engine and LLMs. It is used to build logical reasoning and factual Q&A solutions for professional domain knowledge bases. It can effectively overcome the shortcomings of the traditional RAG vector similarity calculation model.

Seamless Integration Across Knowledge Graphs, Bridging Data Silos

By employing a standardized semantic framework, it becomes feasible to connect diverse, heterogeneous, sequential and intricately related data sources within an enterprise, thereby dismantling data silos

Deep Semantic Contextual Association

By standardizing semantic enrichment of business entity properties, data can be managed knowledge-based, thereby enriching semantic associations among entities and further improving business efficiency

Knowledge Symbolic Representation, Bidirectionally Driven by Large Models

Leveraging a programmable knowledge graph framework makes it easy to combine domain knowledge graphs with large language models(LLM), thereby enhancing the controllability of LLMs in practical applications.