Web日志挖掘及AprioriAll算法的改进

阐述了利用数据挖掘技术对Web日志进行挖掘的可能性及其重要作用.通过用户维的引入,对已有的Apri-oriAll算法进行了改进,改进后的算法降低了原算法的时空复杂度,特别是在新算法中可以对Web日志数据按用户维进行切片,消除了原算法中的I/O相关性.这不仅使对Web日志数据进行的挖掘活动可以分散地进行,而且使多机连网分布式挖掘成为可能,同时还可以实现对Web日志的增量挖掘和动态挖掘.此外,该改进算法的特殊意义还在于它不仅可以将所有用户看作一个整体进行挖掘,而且还可以对不同的用户个体的行为进行独立地挖掘,使挖掘出的结果能够满足对用户个性化使用的需求.

Web log mining and the improvement of the AprioriAll algorithm

This paper sets forth the possibility and importance about applying data mining in Web logs mining.By making use of the user dimension,it offers an improved algorithm based on the original AprioriAll algorithm,which has been used in Web logs mining widely.The improved algorithm not only has lower cost of time and space consuming than the original one,but also eliminates the correlation between the input and output data in the original algorithm by slicing up the Web log as the user dimension.So the mining process can be distributed,and the utilization of the multi-processors is possible.At the same time,slicing by the user dimension,it can make the mining process increasingly and dynamically.Further more,in the improved algorithm,the users can be used as a whole to be mined or be mined individually to fit the personal need.