Large Language Model-based Knowledge Graph construction methods comparison

Knowledge graph construction provides a structured representation of knowledge within a specific domain. It helps to detect entities along with their relationships with each other. There are three major phases in knowledge graphs construction: 1) entity discovery, 2) coreference resolution, and 3) relation extraction. LLMs are playing an important role in all these phases. This comparison will explore the LLM-based knowledge graph construction methods.