Construction Of Knowledge Graph Based On Intellectual Property Cases

Knowledge Graph is a knowledge base that connects structured information through graph structure.The concept of knowledge map was first put forward by Google.Since Google put forward the concept of knowledge map,it has been successfully applied to the search engine,which greatly improves the accuracy of search and makes the search engine more intelligent.Later,all walks of life began to pay attention to the knowledge map and study how to build the knowledge map of the corresponding field.In recent years,with China entering a new stage of development,innovation has become a major driving force leading China’s development.In order to protect and encourage innovation,the state has also issued relevant policies and regulations to protect intellectual property rights.Therefore,exploring the application of knowledge mapping in the field of intellectual property is also a meaningful research point,but the research in this field at home and abroad is still relatively limited.This paper aims to study the construction of knowledge map based on intellectual property cases.The specific work contents are as follows:First of all,through obtaining the first instance civil judgment text related to intellectual property disputes published on China’s online judgment documents,this paper analyzes the judgment text and constructs the intellectual property case schema.Secondly,the algorithm of named entity recognition is analyzed,and the model of BILSTM-CRF is used to recognize the named entity of judgment text;Then,according to the structural characteristics of the judgment text,the rule extractor is constructed,and the rule-based entity relation extraction method is used to extract the entity relation triples in the first and last information of the judgment.Then,the effect of the dependency parsing model in LTP on the entity relation triples in the body of the judgment is analyzed,The model is used to extract the entity relationship triples in the body of the judgment;After,the effect of information extraction model based on dgcnn and probability graph is analyzed,and the entity attribute triples in the judgment are extracted by DGCNN and probability graph information extraction model combined with remote supervision prior features.The last,the extracted knowledge is stored in the Neo4 j graph database to complete the construction of knowledge map,and a knowledge map system based on intellectual property cases is constructed by using vis.js visualization library,Django framework and bootstrap framework for users to query knowledge.It is of great research value and commercial value to build a knowledge map based on intellectual property cases.For ordinary users,it can facilitate users to understand the case information related to intellectual property and improve their awareness of intellectual property protection.For professionals,it can assist them in case study,decision analysis,case reasoning and so on.