基于知识图谱检索增强生成的就业指导问答系统OA

This research proposes an employment guidance Question-and-Answer(Q&A)system based on Knowledge Graph Retrieval-Augmented Generation(RAG).The system achieves intelligent career guidance Q&A through three key technological innovations.Firstly,it automatically constructs a structured career Knowledge Graph using Large Language Models.Secondly,it introduces a two-step hybrid strategy of"mapping transformation followed by categorical alignment"to calculate entity similarity and merge similar entities,effectively addressing issues of"homonymy"and"synonymy".Finally,it integrates structured knowledge from the knowledge graph into the generation process of Large Language Model via RAG,thereby optimizing answers with knowledge-driven enhancements.Experimental results demonstrate that the system significantly improves the professionalism and reliability in career guidance across areas such as job understanding,skill requirements,and career development.This research provides a new technical pathway and practical paradigm for building an intelligent career service system.