摘要
通过聚类方法分析用户的行为模式(喜好和习惯),对Web日志数据进行数据净化、用户识别、会话识别、帧页面识别和路径补充,然后自动生成动态的链接,让用户能迅速地从数量庞大的网站网页中找到自己喜欢的、想要的数据,从而节约了用户的网络资源和时间,也提升了网站的竞争能力。
Through clustering approach analyses user's behavior pattern (hobby and custom), carring on data purification, user recognition,conversation recognition, frame page recognition and way supplement to the Web log data, the automatic producted dynamic link,enables the user to be rapidly found from the quantity huge website homepage valuable data, thus saves user's network resources and time,thus also promotes the website competitive power.
出处
《现代电子技术》
2006年第21期89-91,共3页
Modern Electronics Technique