提高软件Agent策略生成的语义互操作能力  

Improve the Semantic Interoperability for Software Agents Strategy Generation

作　　者：李卫华[1] 杨春燕[2] 方卓君[1] 

机构地区：[1]广东工业大学计算机学院,广东广州510006 [2]广东工业大学可拓工程研究所,广东广州510090

出　　处：《数学的实践与认识》2012年第11期131-138,共8页Mathematics in Practice and Theory

基　　金：广东省自然科学基金(8151009001000041,10151009001000044)

摘　　要：软件Agent遇到语义二义性时无法正确地为用户解决矛盾问题.利用本体支持语义互操作的特点,在软件Agent策略生成机制中引入复合元本体,加入Agent智能引导的人机交互方法,实现了具有语义歧义消除能力的可拓策略生成系统.当用户输入的问题信息语义模糊时,Agent根据本体中的知识与用户交互,逐步理解用户的信息语义,直到能为用户生成解决矛盾问题的策略.以旅游与购物问题为例进行了实验,结果表明了软件Agent策略生成的语义互操作能力得到了提高.When encountered semantics mismatch, software agents cannot solve contradiction problems for users correctly. Take advantage of ontology supporting semantic interoperability, complex-element ontology is built in software agent＇s strategy generation mechanisms. With agent intelligent introduction human-machine interactive method, an extension strategy generation system with semantics mismatch eliminate ability has been built. If users＇ input problem information is semantic obscured agents interact with users according to ontology knowledge and progressively understand user information semantics. Then they can generate strategies for users to solve contradiction problems. Experiments for travel and shopping have been done. The results show the semantic interoperability for software agent＇s strategy generation is improved.

关 键 词：AGENT 策略生成 语义 互操作 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391.1[自动化与计算机技术—控制科学与工程]