Domain Knowledge Aggregation Model Based on Knowledge Graph and Large Language Model

Traditional knowledge aggregation methods are mostly limited to the physical splicing of data and information resources, lacking deep semantic correlations. Domain knowledge aggregation has emerged as a crucial research direction to enhance knowledge organization, management, and services in specialized fields. This study constructs a graph-model-driven domain knowledge aggregation framework based on knowledge graphs and large language models. By integrating the semantic computing capabilities of knowledge graphs with the natural language generation capabilities of large language models, we propose a multi-layered model encompassing knowledge resource collection, semantic association mining, text generation, and service interface integration. The study reveals that the proposed framework lays a foundation for a new paradigm in domain knowledge aggregation. The integration of knowledge graphs and large language models enhances the comprehensibility and reliability of aggregated knowledge, supporting the development of domain knowledge-sharing service platforms and open-service engines.