一种基于自组织映射神经网络的Web页面个性化推荐模型  

AN INDIVIDUALIZED WEB PAGE RECOMMENDATION MODEL BASED ON SELF-ORGANIZING MAP

作　　者：宋庆伟[1] 向阳[1] 

机构地区：[1]同济大学电信学院,上海200092

出　　处：《计算机应用与软件》2007年第2期19-22,共4页Computer Applications and Software

基　　金：国家自然科学基金项目(70371054)。

摘　　要：人工神经网络(ANN)为Web挖掘技术提供了一种新思路。针对传统聚类方法面对大量动态增长的Web日志数据健壮性和灵活性不够的缺点,提出了一种基于自组织映射神经网络的用户信息聚类模型,最后给出的实例证明在此基础上的Web页面的个性化推荐算法能够有效地进行Web页面个性化推荐。Artificial neural network,also known as ANN,has provided a new approach for Web mining. Because traditional clustering method does not perform well in terms of robustness and flexibility when large amount of dynamic Web log data is input, a model which is based on self-organizing map（ SOM）, is introduced to cluster visitor＇s information. And according to the results of experiments, the individualized Web page recommendation service based on that model can work effectively.

关 键 词：人工神经网络 自组织神经网络 WEB挖掘 聚类 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]