基于站点结构的用户频繁路径挖掘  

MINING USER FREQUENT PATHS BASED ON WEB SITE STRUCTURE

作　　者：王琼[1] 刘珏[2] 徐汀荣[1] 

机构地区：[1]苏州大学计算机科学与技术学院,江苏苏州215006 [2]应天学院计算机与信息技术系,江苏南京210046

出　　处：《计算机应用与软件》2009年第1期150-152,共3页Computer Applications and Software

摘　　要：在分析现有挖掘用户频繁路径技术不足的基础上提出算法MUFPS(Mining User Frequent Paths based on Supportability)。该算法主要通过用户会话文件计算出所有被请求页面各自的支持度总和,并结合Web站点结构挖掘出该用户的频繁访问路径。通过实验对比证明该算法能有效地提高挖掘效率,同时保证了挖掘结果的准确性和可靠性。The paper proposes an algorithm called MUFPS （ Mining User Frequent Paths based on Supportability） by analyzing the deficiency of existing techniques of mining user frequent paths. According to the user session file, the algorithm mainly computes the sum of the support degree of every requested page, and mines the users frequent paths in combination with web site structure. The result shows that the algorithm can effectually improve the mining efficiency through the comparison experiment, and ensure the reliability arid veracity of the mining results.

关 键 词：WEB日志挖掘 网站结构 频繁路径 数据挖掘 支持度 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论] TP311.5[自动化与计算机技术—计算机科学与技术]