【作者】 谭小球； 姚敏； 顾沈明；

【Author】 TAN Xiao-qiu, YAO Min, GU Shen-ming (College of Information, Zhejiang Ocean University, Zhoushan 16000 China) (College of Computer Science, Zhejiang University, Hangzhou 310027 China)

【机构】 浙江海洋学院信息学院； 浙江大学计算机学院；

【摘要】 提出一种基于最大频繁序列模式的页面推荐技术,由于考虑了用户会话的页面访问顺序,比一些不考虑页面访问顺序的推荐技术有更高的准确率。通过引入一树型结构,其上压缩存储了所有最大频繁序列,由于前缀相同的序列共享共同的树结点,从而大大节省了存储空间。推荐引擎截取用户活动会话中最近被访问的页面子序列, 与树的部分路径进行匹配,无需在整个模式库中搜索相同或相似的模式,加快模武匹配的速度,更好地满足页面推荐的实时要求。实验证明,方法是有效的。更多还原

【Abstract】 A technology based on maximal frequent sequential pattern is presented to perform personalization recommendation. Because of taking the access sequence of pageviews into account in this technology, a higher accuracy can be obtained. A new tree structure is introduced to compress and store all maximal frequent sequences discovered from Web usage mining, in which the sequences owning common prefixes share same tree nodes so that the storage can be saved greatly. Recommendation engine intercepts nearest access subsequence form user active session to match some sub paths of the tree, without need to find the identical or similar patterns in the whole pattern database, which enormously improved the speed of patterns matching to sadsfy the need of real-time recommendation better. The results of experiments show the approach is effective in speed and precision.更多还原