基于URL相似度的会话识别方法  

Session Identification Method Based on Similarity URL

作　　者：周松松[1] 马建红[1] 

机构地区：[1]河北工业大学计算机科学与软件学院,天津300401

出　　处：《计算机系统应用》2014年第12期191-196,共6页Computer Systems & Applications

摘　　要：随着互联网的快速发展,Web日志的用户行为分析已经成为互联网技术领域的研究热点之一.会话识别是Web日志用户行为分析的关键步骤,精准的会话识别是有效进行用户行为分析的基础.本文在IIS Web日志分析的基础上,提出了一种基于URL相似度的会话识别方法.实验结果表明,此方法可以有效的识别出用户的真实会话.With the rapid development of Internet, Web log user behavior analysis has become one of research hotspots in the field of Internet technology. Session identification is the key step in the Web log user behavior analysis. Accurate session identification is the foundation of effective user behavior analysis. IIS Web log analysis is presented in this paper on the basis of a session identification method based on known URL. The experimental results show that this method can effectively identify the user＇s real conversation.

关 键 词：数据预处理 用户识别 会话识别 编辑距离 最长公共子序列 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]