摘要
现实中存在着许多不能完全表示或不断变化的知识,这就要求语义Web不仅能够处理静态知识,还要对动态知识进行处理。仅利用传统描述逻辑已不足以有效表示相关知识,还需增加对动作的表示和推理。针对以上问题,提出的动态非单调语义Web规则语言(dynamic non-monotonic semantic Web rule language,D-NSWRL),它使用一种新的标记语言表示动态和非单调问题,以描述逻辑为主框架,对动态描述逻辑(dynamic description logic)和非单调逻辑进行整合并加以改进;通过定义清晰的语法和语义,在计算规则中提出动作相关度影响结果因子的概念,解决规则间的冲突问题,以表示规则间的优先顺序,对推定结论的优先序做出合乎直观的判断。将D-NSWRL应用到疾病诊断推理原型系统中,取得了良好的反馈,为语义Web环境下对动态以及非单调领域知识的刻画和推理提供了进一步的逻辑支持。
There is much knowledge that cannot be fully expressed or constantly changes in reality.This requires semantic Web to be able to handle not only static knowledge,but also dynamic knowledge.The use of traditional description logic is not sufficient to effectively represent relevant knowledge.It needs to increase the representation and reasoning of actions.In response to the above situation,the dynamic non-monotonic semantic Web rule language(D-NSWRL)is proposed in this paper,which uses a new markup language to represent dynamic and non-monotonic problems.Firstly,the dynamic description logic(DDL)and non-monotonic logic are integrated and improved with the description logic as the main framework.Secondly,conflicts between rules are resolved by defining clear syntax and semantics.In the calculation rules,the concept of action correlation result factor is proposed to represent the priority order among rules,and the judgment of the presumption conclusion is made intuitively.Finally,the D-NSWRL is applied to the disease diagnosis reasoning prototype system,which obtains good feedback and provides further logical support for the characterization and reasoning of dynamic and non-monotonic domain knowledge in the semantic Web environment.
作者
王星
李佳
陈吉
WANG Xing;LI Jia;CHEN Ji(School of Electronics and Information Engineering,Liaoning Technical University,Huludao 125105,P.R.China)
出处
《重庆邮电大学学报（自然科学版）》
CSCD
北大核心
2020年第1期154-163,共10页
Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)
基金
国家自然科学基金(61402212)
辽宁省高等学校杰出青年学者成长计划(LJQ2015045)
辽宁省自然科学基金面上项目(2015020098)
中国博士后基金面上项目(2016M591452)~~
