基于模糊兴趣集和智能Agent的推荐系统模型研究  

Model of recommender system based on fuzzy interest set and intelligence agent

作　　者：王卫平[1] 赵明[1] 

机构地区：[1]中国科学技术大学管理学院,安徽合肥230026

出　　处：《信息化纵横》2009年第17期59-62,共4页

摘　　要：提出了一种结合用户模糊兴趣和智能多Agent的电子商务推荐系统模型,利用智能Agent获取用户对产品属性的喜好度和相应权重,依此建立起清晰的用户兴趣模型。在智能多Agent技术下建立一种多属性综合评价机制,采用逼近理想解法(Topsis法)对候选产品进行实时处理,并根据排序结果产生Top-N推荐。The paper proposes a model of electronic business recommendation system based on users＇ fuzzy interest set and intelligence multi-agent. By using intelligence agent, user preferences and its weights for product attributes are obtained, and a clear model of user interest established. The model also applies multi-agent to build a comprehensive multi-attributes evaluation mechanism. The topsis method has been used to perform real-time ranking. After that, the Top-N are recommended in accordance with the results.

关 键 词：推荐系统 智能AGENT 模糊兴趣集 多属性决策 逼近理想解法 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]