Font Size: a A A

Research On The Influence Of Dependency Syntax On The Construction Quality Of Domain Knowledge Graph

Posted on:2022-01-28Degree:MasterType:Thesis
Country:ChinaCandidate:H FanFull Text:PDF
GTID:2517306722478784Subject:Education Technology
Abstract/Summary:PDF Full Text Request
In the era of big data,technology is changing with each passing day.The resources supported by the Internet growing exponentially explodes to users.Faced with such a large amount of data,even the indigenous people of the data still have certain difficulties in data acquisition and processing.Nowadays,It is the top priority to select data quickly and effectively,and organize target data to serve one’s own needs.As a relatively special field,the audiences of basic education field are teachers and students.Most learners do not have enough ability to quickly obtain the content they need from massive data.In other fields,the resources in this field are sorted through the construction of knowledge graphs,and the system is integrated into a domain knowledge base to provide great help for search engines,domain knowledge question and answer,and semantic search.Some people in the field of basic education also try to construct a subject knowledge graph to sort out the knowledge structure,facilitate knowledge retrieval and knowledge question and answer.However,there are some problems in the construction of knowledge graph in the field of basic education,mainly focusing on the choice of construction technology and the quality of construction.The specific problems are:(1)The validity and applicability of the technology used in the knowledge map in the field of basic education need to be considered.Although there are a large number of technologies that can achieve the purpose of building a knowledge map,no one has studied the effectiveness and applicability of the technology in depth.The characteristics of various disciplines in the basic education field are relatively clear.If you can study a set of knowledge map construction systems applicable to various disciplines and technology will greatly reduce the difficulty of constructing knowledge graphs.(2)The quality level of knowledge graph construction is uneven.There are already more authoritative universities that construct a knowledge map of subjects in the field of basic education,but the quality of the construction has not been studied in depth.His research only set up some hypothetical conditions for the construction of the knowledge graph in the early stage of construction,and the premise of the knowledge graph applied in various fields is that its authority can be guaranteed,and its quality needs to withstand consideration.It seems that there is no in-depth analysis of this..It is embodied in that the domain knowledge graph is not strong in domain,comprehensive,and accurate.It analyzes the typical methods in the two major links of knowledge graph construction,combines the advantages of dependency syntax in sentence structure,and assumes that dependency syntax has a positive impact on the quality of knowledge graph construction in this study.In order to verify the hypothesis,four comparative experiments were designed for the two major processes of the knowledge map construction,and the specific impact on the construction process and the statistical analysis of the construction result data.In order to get a more accurate conclusion,the article quantified the domain,accuracy,coverage and comprehensiveness of the entity extraction,entity expansion,entity relationship extraction and the final construction of the knowledge graph.Finally it found that the dependency syntax is in the knowledge graph construction has advantages in quality.Ans the study has achieved good results in entity extraction and entity relationship extraction.In the future,it can be applied to the construction of knowledge graphs in other fields to provide a certain reference for the quality of knowledge graphs in other fields.
Keywords/Search Tags:Dependency syntax, Named entity recognition, Relation extraction, quality evaluation of knowledge graph
PDF Full Text Request
Related items