基于隐马尔可夫模型的兴趣迁移模式发现  

Mining Interest Navigation Patterns Based on Hidden Markov Model

作　　者：王实[1] 高文[1] 李锦涛[1] 黄铁军[1] 

机构地区：[1]中国科学院计算技术研究所,北京100080

出　　处：《计算机学报》2001年第2期152-157,共6页Chinese Journal of Computers

摘　　要：Web挖掘的一个重要研究方向是发现用户的迁移模式 .一般来说 ,用户的迁移具有某种目的性 .这种目的性表现为用户对某种概念的兴趣 .文中提出基于隐马尔可夫模型的兴趣迁移模式发现方法 ,用于发现这种带有某种兴趣的用户迁移模式 .这种模式实质上是一种特殊的关联规则 .在这种方法中 ,作者首先根据用户的访问记录定义一个隐马尔可夫模型 ,然后提出一种新的增量发现算法 Increase- R用于发现兴趣迁移模式 ,同时给出了证明以说明该算法可以发现所有的兴趣迁移模式 .Mining Navigation patterns is an important research direction in web mining. The discovered Navigation patterns can be used to help the designers to understand the users' access actions, improve the structure design, carry out the advertisement, and get the users' characteristics. In general, a user accesses a web site with some intentions. These intentions represent the interest in some conceptions. So the user's interest has some relation with his navigation path. The users' interest navigation paths compose the users' interest navigation patterns. In this paper, we present a new method for mining interest navigation patterns based on the hidden Markov model in order to discover users' interest navigation patterns. These patterns are a kind of the special association rules essentially. In our approach, we build a hidden Markov model according to web server logs firstly, then we present a new incremental discovery algorithm Increase _R in order to discover the interest navigation patterns, and we testify that the algorithm can find all interest navigation patterns.

关 键 词：数据挖掘 隐马尔可夫模型 关联规则 数据库 兴趣迁移模式 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论] TP18[自动化与计算机技术—计算机科学与技术]