Research On Linked Data Quality Assessment Method Based On Ontology

With the continuous development of semantic web technology,the semantic data of the network presents explosive growth.With the development of knowledge Graph,the research and application of network semantic data are more common.As the carrier of the storage and exchange of the semantic data of the network,the number of publish and use of the Linked Data also increase exponentially.Because of the openness and sharing of the network,many units publish their data in their respective fields to the network through the Linked Data technology for everyone to use.However,the standards that data publishers follow are not uniform,and the data quality of published Linked Data is difficult to guarantee,which makes the data quality of Linked Data become the key factor that hinders the application of Linked Data.In order to reduce the impact of Linked Data quality problems,many enterprises and units are looking for effective methods to assess the quality of Linked Data,many scholars at home and abroad have also joined in the research of the quality of Linked Data.According to the research on the quality of Linked Data,some experts think that the quality of Linked Data is essentially the quality of data,while some scholars regard the quality of Linked Data as a research direction alone.But up to now,there is no standardized method to evaluate the quality of Linked Data.However,most of the domestic scholars are more focused on the statistical analysis of the quality assessment methods of the Linked Data.In order to solve the problems of low efficiency and low integration of methods in the quality assessment of Linked Data.First of all,this thesis studies the related thesis of Linked Data and Linked Data quality,and analyzes the research results at home and abroad.The necessary dimensions of Linked Data quality are discussed,and the detection methods of Linked Data quality are compared.Secondly,this thesis studies the network data transmission technology to provide technical support for the evaluation of the exchange and transmission efficiency of Linked Data.It can complete the detection of Linked Data delay,throughput and other aspects.Thirdly,the current Linked Data quality assessment model is not standardized.This thesis studies and analyzes the data quality assessment model and the Linked Data quality assessment model,then uses Jena to build the LDQA ontology for the Linked Data quality assessment according to Seven-step.LDQA can standardize the concept of Linked Data quality,and it can use its reasoning function to discover and mine the hidden knowledge,and improve the accuracy of assessment.Based on the oilfield development data,the Linked Data quality assessment system is designed and implemented.The system integrates assessment methods of different dimensions of Linked Data Quality.The maintenance interface of assessment ontology is provided.It can maintain and upgrade ontology and reasoning rules.And the operation access of users and managers is separated,which reduces the complexity of the system.The quality of Linked Data is assessed efficiently and accurately.The validity and feasibility of the method based on ontology are verified.