摘要
人在信息活动中所具有的复杂性决定了大数据时代下人物信息资源的重要程度,因此挖掘人物关系是提高信息资源质量、构建个人实体知识网络、为用户提供知识服务的有效途径。本文将跨领域的因果链求解算法运用到人物亲属关系的挖掘问题中,并引入罗马亲等计算法衡量不同类型的亲属关系远近,定量地描述了个人实体之间的关联程度,同时采用模糊认知图作为人物关系知识表达模型,最后以Wikidata知识库为实验数据来源,进一步验证该算法的有效性,同时将亲属关联权重值开创性地添加到家族亲属关系图谱的绘制中,并构建了个人实体知识关联模型。本文旨在进一步优化网络环境下的人物信息,实现基于人物关系的知识关联,以期为揭示全方位的人物关系的表示方法提供新视角。
In the era of Big Data, interpersonal information resources is becoming increasingly important because of the complexity of people in the information activities, thus mining interpersonal relationship is the effective way to improve the quality of information resources, to build personal knowledge network and to provide users with infor- mation services. In this paper, we apply the cause-and-effect chains algorithm to the problem of mining family rela- tionship, then we introduce the Degree or Kinship in Roman Law to identify the different types of kinship, to measure the degree of kinship between the individual entities, while using the FCM to building the knowledge representation model of character, and finally select the experimental data in Wikidata, to further validate the algorithm, while draw maps of family kinship adding the relative weights, building knowledge model of personal entity. This article aims to further the quality of character information in network environment, achieving the associated character relationships, in order to provide a new oerst)ective to reveal the full range of representation of the interpersonal relationship.
作者
贾君枝
冯婕
Jia Junzhi Feng Jie(School of Economics and Management, Shanxi University, Taiyuan 03000)
出处
《情报学报》
CSSCI
CSCD
北大核心
2017年第3期221-230,共10页
Journal of the China Society for Scientific and Technical Information
基金
国家社会科学基金重点项目"基于关联数据的中文名称规范档语义描述及数据聚合研究"(15ATQ004)
关键词
人物关系挖掘
亲属关系
因果链算法
Wikidata
interpersonal relationships mining
family relation
algorithm of detecting cause-and-effect chains
Wikidata