Research On The Construction And Application Of Military Terminology Knowledge Graph

Military terminology is a professional language in the military field.It is a normative and standardized language for soldiers.It is the basic requirement for unifying military cognition and the basic guarantee for coordinating military actions.With the continuous development of military technology and theory,new military concepts and their referent forms emerge in an endless stream.The nominalization and standardization of these new concepts and referents need to be updated urgently,and there is an urgent need for innovative military language management techniques and methods.Due to the reasons such as not timely updating of military terminologies,not high degree of standardization,backward technology and methods,it is common to find inconsistent semantic reference in the application of military terminologies,and the logic of military-related articles is not strict.First,the semantic correlation between military terminologies is weak,which is difficult to apply to military intelligence and other related applications,and it is difficult for users to intuitively and comprehensively understand the concept of military terminologies and their semantic correlation.Second,most military terminologies exist in paper form,which makes it inconvenient to inquire,popularize and apply military terminologies.Therefore,this paper constructs military terminology ontology,knowledge graph and its application system for military terminology knowledge,so as to provide technical and methodological support for uniform semantics and standardized application of military terminology.(1)Construction of Military Terminology Knowledge GraphThe military terminology dataset consists of internal printed books and Chinese military encyclopedias.Under the guidance of experts in the military field,this paper has an in-depth analysis of the vocabulary,concept and relationship of military terminologies.The Protege ontology modeling tool is used to build the military terminology ontology,and the consistency of the ontology structure is analyzed by using the Hermi T reasoning tool,which provides the top-level design pattern for the construction of the military terminology knowledge diagram.By describing the knowledge system of military terminologies and analyzing the data of military terminologies,this paper constructs and identifies the entities of military terminologies in a template-based way,and provides the entity information description for the subsequent extraction of military terminologies.Finally,this paper uses the rule-based way to extract the military terminologies with significant features of semantic association relations,and initially constructs the knowledge graph of military terminologies.At the same time,the semantic relation of military terminologies extracted by the rule-based method also provides training data for the military terminologies relation extraction model.By fine-tuning the BERT-wwm pretraining model,the model constructed in this paper achieves the best results in the military terminology data set,and the effectiveness of the model structure is verified by ablation experiments.The combination of the above two relation extraction methods provides an accurate and efficient extraction tool for military terminologies relation extraction.(2)Application of Military Terminology Knowledge GraphThe military terminology knowledge graph consists of nodes representing the entity information of military terminologies and edges representing the semantic association between military terminologies.In order to facilitate the management and promotion of military terminology knowledge graph,this paper constructs a military terminology knowledge service system,which provides users with a highly available knowledge service system by separating the front and back ends.The server uses Neo4 j to store the military terminology knowledge graph,and uses Spring Boot framework to uniformly process the request information sent by the front-end user.The front-end interface adopts VUE architecture to provide users with a convenient and quick interface for knowledge management,inquiry and analysis of military terminologies.To sum up,this paper systematically integrates and analyzes military terminology data,and uses Protege modeling tool to build military terminology ontology,providing an accurate and comprehensive knowledge system for military terminology knowledge sharing and reasoning.According to the characteristics of military terminologies data,this paper proposes a military terminologies entity recognition method based on template,which can efficiently identify the entity and attribute information in the corpus data of military terminologies,and provide high-quality entity information for the construction of military terminologies knowledge graph.Secondly,this paper proposes two combined relationship extraction methods based on rules and deep learning model.The combination of the two methods enhances the semantic correlation between military terminologies and provides an efficient extraction tool for military terminologies relation extraction.Finally,this paper constructs a military terminology knowledge service system on the military terminology knowledge graph,which provides an efficient and fast service platform for the standardization and popularization of military terminology.