Research On The Personal Information Aggregation Of The Chinese Name Authority File

The main work of library authority control include the establishment and management of name authority files.In 2003,a number of institutions in Mainland China,Hong Kong,Macau,and Taiwan jointly established Chinese Name Authority Joint Database,and provided users with a onestop service for querying.However,the management of the data resources by different institutions lacks a standard integration model and is poorly shared,the data source is relatively single,which reduces the service efficiency and information quality of the database.Person is the most important entity in the objective world,the related data resources are highly redundant in the Chinese name authority file,and multiple items of the same person lack semantic association and information integration.It is significance to complete the information of personal entities,and then improve the data quality of library resources.At present,many domestic and foreign knowledge bases cover multiple types of entity data such as people,organizations,and conferences,each knowledge base has its own characteristics in the way of constructing entity objects,describing attributes,and semantic associations.The Chinese name authority file lacks attributes,and the format is difficult to link.Information aggregation has become an effective way to solve the above problems,it can integrate duplicate entries from multiple domestic institutions,provide directions for identifying the same entity,and establish links with other types of databases.This paper has conducted theoretical and empirical research based on the information aggregation of people in Chinese name authority files.It mainly includes the following aspects:(1)The paper analyzes the problems of multi-agency data aggregation in domestic name authority files.For personal entries scattered,proposed clustering method to integrate personal data resources;(2)Explore ways to match entity with virtual international authority files,analyze how organizations around the world match authority names,design database entity matching experiments,evaluate the effectiveness of the matching of person entities based on name attributes,and establish a basis for the integration of domestic normative files with international standards and data sharing;(3)Analyze the content of the attributes described by Wikidata entities,use the vector space model to measure the similarity of the Chinese Name Authority Joint Database.According to the standard principles of the Semantic Web,it will be published as Linked Data,which will expand the use of name resources,overcome heterogeneous and further realize the sharing and use of library resources.