基于关联规则的Wikidata人物名称数据分析——以诺贝尔文学奖得主为主题

Data Analysis of Wikidata Person Names Based on Association Rules Mining:A Case Study of the Theme of the Nobel Prize Winner

[目的 /意义]挖掘不同名称数据之间的关联关系,将关于某一实体或主题的领域知识表现出来,这对实现不同层次、不同粒度的知识体系的解构和重构、提供满足多种需求的知识服务工作具有重要的研究意义。[方法/过程]提出一种基于人物实体数据运行关联规则挖掘实验的研究框架,通过对人物实体条目的抽取、预处理及属性识别与分类等处理方法,利用R语言得到人物实体集的关联规则,实现多种名称数据的关联,最后从Wikidata知识库提取113位诺贝尔文学奖得主的实体条目进行实证分析。[结果 /结论]分析右部为地点名称、机构名称、时间名称和主题名称等4种不同类型规则的关联特征,实现不同名称数据类型的关系挖掘问题。本研究可为知识的揭示、聚合和关联提供新的视角,探索了数据挖掘技术在名称数据中的应用。

：[Purpose/significance] Mining the relationship among different name data to show a domain or subject knowledge of a particular entity, which to achieve different levels, different dimensions of the knowledge system deconstruction and reconstruction, to provide a variety of needs to meet the knowledge service work has important research significance.[Method/process] This paper presents a research framework based on the association experiment of association rules of character entity operation. Through the extraction of object entity entries, preprocessing and attribute recognition and classification, the paper uses R to get the association of human entity rules, to achieve a variety of name data association, and finally extracts 113 Nobel Prize winner entity entries from the Wikidata knowledge base for empirical analysis.[Result/conclusion] The relationship between four different types of rules, such as place name, institution name, time name and subject name, is analyzed, and the relationship mining problem of different name data types is realized. This study provides a new perspective for knowledge disclosure, aggregation and association, and explores the application of data mining technology in name data.