摘要
分析了传统搜索引擎所存在的问题,提出了一种面向用户的智能搜索引擎模型(User-Oriented Intelligence Search Engine,UOISE)和动态Q学习算法,UOISE模型由若干个Agent和知识库组成,针对用户个人信息以及用户的上网行为,分析用户的兴趣所在,从而推断出用户的意图,能有效地为用户提供真正需要的、即时的信息,为用户提供多层次个性化的信息服务。动态Q学习算法实现了Agent跟踪用户行为、推断用户兴趣的过程。对比实验表明,UOISE比普通检索工具具有更高的查全率和查准率。
The problems existing in traditional search engine are analyzed in this paper.The model of User-Oriented Intelligence Search Engine,UOISE,and the dynamic Q learning algorithm are put forward.The model consists of the muhi-agent systems and the rule base.According to the user＇s information and the user＇s action when he/she browses the web,UOISE obtains user＇s intention,and serves him/her,the personal information.The dynamic Q learning algorithm realizes the process that the agents track user＇s action and deduce user＇s interesting.Compared with the traditional search tool,UOISE has more all-searching probability and more accuracy.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第35期175-177,232,共4页
Computer Engineering and Applications
基金
江苏省研究生创新基金(xm-35)。
关键词
智能搜索引擎
多AGENT系统
动态Q学习
兴趣度
intelligence search engine
multi-agent systems
dynamic Q learning algorithm
user＇s interest
