Design And Implementation Of Military Equipment Question Answering System Based On Knowledge Graph

In the era of big data,information on the Internet is increasing day by day,and people have put forward higher requirements for information acquisition.In the context of the digitization of military command and the continuous advancement of the modernization of military equipment,how to effectively refine,manage,and use military information is of great significance for improving the application of military knowledge and the efficiency of military resources.Therefore,this project takes the field of military equipment as a research entry point,and builds an intelligent question answering system for military equipment based on knowledge graph technology.The main work of this project includes:(1)Collect and analyze the semi-structured data of the military weapons website,construct the military equipment knowledge graph ontology,standardize the field information of the data,and process the data into entities,attributes and attribute values,as well as entities,relationships,and entities in the form of triples.import the triple data into Neo4 j to construct a knowledge graph of military equipment.(2)The knowledge graph question answering is divided into two main sub-processes,named entity recognition and attribute linking.Use bidirectional long and short-term memory network(BiLSTM)and conditional random field(CRF)to complete the task of named entity recognition.In order to solve the traditional entity recognition algorithm using static word vectors,which cannot represent the semantics of the same word in different contexts,the BERT language model is adopted to encode word vectors,apply the BERT-BiLSTM-CRF model to the process of military equipment named entity recognition.The attribute linking process is transformed into a sentence-pair classification problem,and BERT sentence-pair classification fine-tuning model is used to solve this problem.Experiments on these two sub-processes respectively show that the algorithm used in this paper can accomplish the task well.(3)Designed and implemented a question answer system for military equipment based on knowledge graph.On the basis of the constructed knowledge graph in the field of military equipment,combined with related intelligent question answering algorithms,a military equipment question answer system based on the knowledge graph is realized,which can answer users’ questions in the field of military equipment.