基于模糊等价关系的文本多粒度划分方法  

The Method of Dividing Documents into Multi-level Granules Based on Fuzzy Equivalence Relationship

作　　者：刘平峰[1,2] 余文艳[2] 游怀杰[1] 

机构地区：[1]武汉理工大学经济学院武汉430070 ． [2]武汉理工大学信息工程学院武汉430070

出　　处：《情报学报》2012年第6期589-594,共6页Journal of the China Society for Scientific and Technical Information

基　　金：教育部人文社科项目“面向多粒度知识服务的Web信息融合研究”（10YJAZH050）.

摘　　要：由于知识的粒度性反映了人类认识世界的特征，本文提出一种基于模糊等价关系的文本多粒度划分方法，以模糊等价关系构建文本信息颗粒，通过模糊等价关系入截集阈值的控制得到信息的文本多级粒度划分，进一步结合聚类主题词识别方法确定各级信息颗粒主题词之间的包含关系，为多粒度知识服务基于主题的导航提供基础。实验结果证明了该方法的有效性。Granularity of knowledge reflects the characteristics of the human being＇s understanding of the world. This paper presents a method of dividing documents into multi level granules based on fuzzy equivalence relationship. Granules are formed according to fuzzy equivalence relation. Then multi-level granules are formed by controlling the value of threshold kof cut sets based on the fuzzy equivalence relation. After the theme of each granule is discerned, the inclusion relations between the themes of granules located at the adjacent levels are derived. The derived results can be exploited for the theme based navigation for multi-granular knowledge service. The experiment demonstrates the effectiveness of the above method.

关 键 词：模糊等价关系 粒计算 文本相似度 

分 类 号：G253[文化科学—图书馆学]