Design of Question Answering System for Railway Electricity Accidents

摘要

Based on multi-source data of railway electrical accident, an intelligent Q&A(question answering) system based on knowledge graph of railway electrical accident is realized in this paper, which can improve the intelligence and convenience of accident information retrieval. First of all, in order to complete the construction of the railway electrical accident knowledge graph, the conceptual structure design, knowledge extraction, and knowledge storage technology are used in this paper. In addition, the overall architecture of the railway electrical accident intelligent Q&A system is designed, and question preprocessing, question classification, and answer retrieval techniques are used to achieve the Q&A system. Finally, in order to achieve the functions and visualizations of Q&A system, this system is built based on Flask-framework, combining Echarts module, data tables and Neo4j-knowlegde-graph. Therefore, correctness, scientificalness and visual effects are guaranteed.