Research And Implementation Of Intelligent Question Answering System Based On Knowledge Graph

With the explosive growth of information in the Internet,it is more and more difficult to meet the needs of users for exact answers through the traditional search engine retrieval results.In this context,the question answering system supporting natural language form has become a key research direction at home and abroad.Knowledge graph provides powerful semantic processing ability for knowledge organization and application of Internet content,and effectively solves the problems of loose organization and chaotic structure of Internet content.With the development and practical application of knowledge graph technology,question answering based on knowledge graph has gradually become a hot topic in the field of automatic question answering.The research of question answering technology based on knowledge graph will greatly facilitate people to obtain exact answers from existing knowledge.Based on the knowledge graph,this thesis studies the question answering technology,which includes three aspects: 1.Aiming at the problem of chinese named entity recognition in the open field,this thesis proposes a named entity recognition algorithm based on integrated learning,which can effectively improve the accuracy of named entity recognition.2.Aiming at the problem of relationship extraction which is lack of labeled samples,this thesis proposes a relationship extraction algorithm based on remote supervision and reinforcement learning technology,which can greatly reduce the impact of noise data in the training process and significantly improve the performance of relationship extraction.3.In view of the problem that knowledge items in knowledge graph are often missing seriously,this thesis proposes an improved algorithm based on Path-RNN model,which can infer large multi semantic knowledge data,and support the joint learning and inference of relationship type,entity and entity type.According to the above algorithm,this thesis designs a chinese question answering system based on knowledge graph,which can accurately answer the questions asked by users in natural language.The experiment shows that the system has good performance while running stably.