基于站点结构和浏览时间的路径补全算法

Path completion based on structure of web site and glance time of user

路径补全旨在提高Web使用挖掘的准确性。在网页结构复杂的情况下,两个页面之间可以补充的路径可能有多条,传统的补充方式准确率较低。对此,分析了影响路径选择的因素,提出了用户浏览时间的概念,并结合动态站点结构提出一种基于站点结构和用户浏览时间的路径补全算法。该算法使用有向图结构存储了用户浏览时间和站点结构信息,并根据二者找出最合适的路径。实验结果表明,与传统方法相比,该方法提高了路径补充的准确率。

Path completion is intended to obtain accurate results of Web use mining, There may be many paths between two pa- ges when the Web structure is complex. At this time, traditional way of path completion is inefficient. So the factors are ana- lyzed, which affect path choice, and the concept of browsing time is presented, and a path completion algorithm is proposed based on the structure of Web site and browsing time of user with dynamic site structure. The algorithm can find best path by u- sing browsing time of user and structure of Web site, which are stored in the form of a directed graph. Experiments prove that this algorithm can improve the accuracy of path complement.