基于FCA的Folksonomy用户兴趣研究述评

A Review of User Interest Identification in Folksonomy Based on FCA

传统的Folksonomy用户兴趣研究在相似性计算上更多考虑的是整体相似,不能突出用户之间的差异,同时还存在兴趣结构不符合用户自身认知等问题。形式概念分析从用户自身的认知结构出发,不仅能表示标签间的语义信息,还能形式化标签的结构关系,为用户兴趣的表征和挖掘提供了新的思路。基于FCA的用户兴趣研究较少,主要集中于单用户和用户群兴趣识别两个视角。本文在分析用户兴趣识别传统方法及局限性基础上,进一步探讨了基于FCA进行Folksonomy用户兴趣识别的基础以及现有的相关研究。

The traditional user interest researchs in Folksonomy, often considerate the overall similarity when similarity cal- culation, which can not highlight the differences between users. In addition, there are problems such as interest structure does not meet the user＇s own cognition. Formal concept analysis which based on the user＇s cognitive structure, can not only express the se- mantic information between labels, but also be used to formalize the structural relationships of tags, provieds a new train of thought for user interest＇s characterization and excavation. There aren＇t so many researchs about user interest in Folksonomy based on FCA, which mainly focused on two perspectives： single user and user group interest identification. The paper analysed the tradi- tional methods and their limitations of user interest identifion. And beyond this, this paper also discussed the basis of FCA used in identitlying user interest in Folksonomy and some research about user interest based on FCA.