语义Web下以用户注意力为中心的知识形式化方法  

Attention centered knowledge formalization method in the Semantic Web

作　　者：周旋[1] 李清[1] 周剑[1] 陈禹六[1] 

机构地区：[1]清华大学自动化系,北京100084

出　　处：《清华大学学报（自然科学版）》2007年第10期1591-1594,共4页Journal of Tsinghua University(Science and Technology)

基　　金：国家"八六三"高技术项目(2001AA415340);国家自然科学基金资助项目(60474060)

摘　　要：知识的形式化可以改善知识复用的效率与灵活性。语义Web中现有形式化方法仅把知识看作"结果",追求机器可读可解释性以及推理应用,而忽视用户知识处理"过程",无法广泛应用于一般用户的知识形式化。该文针对一般用户需求,从认知效率角度提出一种以注意力为中心的知识形式化方法。引入信息块和焦点树,实现对注意力的形式化表示、跳转控制和历史记录。比传统方法降低了表示粒度,丰富了显性语义信息,能够满足一般用户对大量经验的管理需求,并为进一步应用提供基础。Knowledge formalization is the foundation for effective utilization of knowledge. Most current semantic web languages focus on knowledge in its ＂result＂ form, aiming at machine readability and interpretability, and logic based reasoning. Such methods, however neglect the ＂process＂ of user knowledge processing, which baffled effective human machine interaction. An analysis of general users＇ requirements was used to develop an attention centered knowledge formalization method from a cognitive efficiency perspective. The method uses of information chunk and an attention tree, to analyze the representation, shift control and history of attention. With lower granularity and richer semantics, it meets the management requirement of the enormous experience of users and its further application.

关 键 词：语义WEB 一般用户 注意力 亚文档 元数据 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]