Intelligent Graph = Knowledge Graph + Intelligent Agents
Recently there has been much excitement related to Artificial Intelligence and Knowledge Graphs, especially regarding the emerging symbiotic relationship between them: LLMs provide unstructured reasoning, whilst the knowledge graph provides complementary structured reasoning. But how do we bridge the unstructured and structured worlds? Does the LLM push or does the KG pull? With LLM-Push we can set up agents which can push information from the LLM into a Knowledge Graph. What data? How often? When do we stop pushing? And so on. With KG-Pull we can add Intelligent agents to a Knowledge Graph, which pulls information from outside, including LLMs. This avoids unnecessarily pushing any data just-in-case into the graph store, instead pulling the data just-in-time when a user questions the graph.