个性化智能信息检索的设计与实现

The Design and Implement of Personalized Intelligent Information Retrieval

文章提出了一个个性化智能信息检索系统NetLooker，它是智能信息检索领域的重要研究课题。系统运用人工智能的方法，采用两层分布式智能体Agent技术、相关反馈学习算法、信息滤波算法，面向用户个性化模式来设计和实现。系统采用两层交互机制，支持个性化检索和浏览式检索两种信息检索方式，有良好的交互方式、能智能适应用户兴趣和信息源的变化。它可应用于WWW、电子商务等分布式系统中进行信息检索，因此具有理论价值和使用价值。

This paper presents an architecture of a personalized intelligent information retrieval system－NetLooker，which is a key problem in the intelligent information retrieval domain．The AI＇s methods，two－tier distributed intelligent agent techniques，the relevant feedback approach and the information filtering algorithm，are used in the system，which is based on the user－personalized models．It employs two－tier collaboration mechanism，supports personalized－and browsing－retrieval．The system has the features of adjusting to users＇ interests and changing information－source adaptively．So it can be used for information retrieval in the distributed systems such as WWW，electronic business，and has high significance in theories and applications．