摘要
用户访问模式发现是构建自适应网站的关键技术.提出了一种基于粗糙集和神经网络相结合的用户访问模式的发现方法,建立了用户访问模式的一般模型.针对Web日志数据通常数据量大、冗余,以及页面之间关系不确定的特点,将粗糙集作为前端预处理器,简化信息处理空间,去掉冗余,采用神经网络聚类分类用户群,从而发现用户访问页面的方式.最后,通过实验分析表明文中方法的有效性.
Discovering of user access pattern is the key technique of adaptive website. In the paper, a method of dectecting user's pattern is built based on the theory of the rough set and kohonen neural network, then the general model of user pattern discovery is established on the basis of rough set and kohonen neural network. Rough set is used as the front pretreatment implement, for it can reduce information process space, get rid of redundant infonnation,and the users can be clustered into many groups. So that, a mode of user access pattern is discovered. In the end, the experiment indicates that the method described in the paper is valid.
出处
《兰州交通大学学报》
CAS
2008年第6期96-99,102,共5页
Journal of Lanzhou Jiaotong University
基金
国家自然科学基金(10771091)
甘肃省自然科学基金(3ZS042-B25-038)
甘肃省教育厅研究生导师科研项目资助(0704-13)