Mining web log sequential patterns with layer coded breadth-first linked WAP-tree

Abstract

Sequential mining is the process of applying data mining techniques to a sequential database for the purposes of discovering the correlation relationships that exist among an ordered list of events. An important application of sequential mining techniques is web usage mining, for mining web log accesses, which the sequences of web page accesses made by different web users over a period of time, through a server, are recorded. Web access pattern tree (WAP-tree) mining is a sequential pattern mining technique for web log access sequences. This paper proposes a more efficient approach for using the BFWAP-tree to mine frequent sequences, which reflects ancestor-descendant relationship of nodes in BFWAP tree directly and efficiently. The proposed algorithm builds the frequent header node links of the original WAP-tree in a Breadth-First fashion and uses the layer code of each node to identify the ancestor-descendant relationships between nodes of the tree. It then, finds each frequent sequential pattern, through progressive Breadth-First sequence search, starting with its first Breadth-First subsequence event. Experiments show huge performance gain over the WAP-tree technique.