Key Technology Research And Application On Ontology Concept Discover And Its Relation Extraction

With the rapid development of network technology,internet has become the platform for people to share and apply the information.How to get valuable information from the huge web data is a big challenge.Semantic-based information processing can effectively solve the above problem.As a shared concept model,ontology pays an important role in the semantic analysis.The research of ontology concept extraction and its relation extraction is an important research topic in natural language process,and it can also extract the related knowledge.The research has important value.Ontology concept extraction and its relation extraction are the most important task on this domain.This thesis does research on ontology concept extraction and its relation extraction.Focusing on electric data,this thesis combines rules and statistics to finish ontology concept extraction.Furthermore,this thesis combines lexical,syntactic and word similarity,to complete ontology relation extraction.The main research work is shown as follows:1)On the basis of the combination of lexical and statistics,this thesis processes the word segmentation.On the basis of the combination of rules and statistics,focusing on the electric data,this thesis completes concept extraction.2)On the basis of word segmentation,part-of-speech tagging,syntactic analysis and the combination of lexical and syntactic features,this thesis constructs the rule base.Furthermore,this thesis does research on ontology taxonomic relation extraction of electric data by the means of related models analysis.3)This thesis adopts the proposed syntactic feature to complete the basic template extraction.Then bootstrapping is used to expand the template.On the basis of that,word vector is used to express concept and the thesis can finish the similarity calculation of triple concept and achieve the non-taxonomic relation extraction of electric data.Experiment result shows that the proposed approach is feasible and effective.The ontology concept extraction method can finish new concept extraction;the precision of ontology taxonomic relation extraction can reach above 40%;word vector is used to finish the similarity calculation.The exiting problems and the next research works are also present in the end.