面向知识服务的用户兴趣动态知识空间模型研究

Model of Dynamic User-Interest Knowledge Space Based on Knowledge Service

目前国内外面向用户兴趣挖掘的研究主要是基于静态研究视角,无法有效反映用户兴趣的时序变化和漂移状态。本文提出在描述用户兴趣的知识关联架构基础上,从用户知识范畴的获取、知识关联挖掘及动态知识演化角度反映用户信息需求的演变,以实现用户兴趣的动态挖掘。通过测度用户兴趣随时间的推移而表现出的知识概念漂移及变化趋势,阐述用户兴趣在各个不同粒度知识单元之间表现出的偏向性及延展性,以提高用户兴趣分析方法的科学性和有效性。

At present, the research for user interest is mainly based on static perspective, which couldn＇t effectively reflect the time and knowledge changes of user interest. This paper aim to describe knowledge association infra- structure, to reflect the perspective of evolution of user information needs from obtain of knowledge boundaries, knowledge mining and dynamic knowledge evolution, to achieve dynamic mining of user interest. By the means of following user interests＇ changes over time and recording the knowledge concept drift and trends, this paper will set forth the bias and ductility of user interest between various knowledge elements, to increase the scientificity and effi- ciency of user interest analysis.