Frequent pattern mining can well extract insight from transaction patterns, and it is a desired capability for fully understanding the customer's purchase behavior. However, most of the algorithms are focus on the transverse relationship and the longitudinal analysis is missed. To address this defect, FP-ICA, a Frequent Pattern mining algorithm for Item-oriented and Customer-oriented Analysis is proposed. A pattern with its items occur in the same transaction is item-oriented, and a pattern with its items occur cross several transactions of a customer is customer-oriented. FP-ICA transforms the transactions to a bitmap which contains a header for recording customer information, and the frequent patterns are obtained by logic And-operation. Different mining rules are used for item-oriented and customer-oriented discovery. Experiments are conducted to demonstrate the fast speed achievement and good scalability of FP-ICA.

Frequent intra-sequence pattern mining and inter-sequence pattern mining are both important ways of association rule mining for different applications. However, most algorithms focus on just one of them, as attempting both is usually inefficient. To address this deficiency, FIIP-BM, a Frequent Intra-sequence and Inter-sequence Pattern mining algorithm using Bitmap with a maxSpan is proposed. FIIP-BM transforms each transaction to a bit vector, adjusts the maximal span according to user's demand and obtains the frequent sequences by logic And-operation. For candidate 2-pattern generation, the subscripts of the joining items should be checked first; the bit vector of the joining item will be left-shifted before calculation if the subscript is not 0. Left alignment rule is used for different bit vector length problems. FIIP-BM can mine both intra-sequence and inter-sequence patterns. Experiments are conducted to demonstrate the computational speed and memory efficiency of the FIIP-BM algorithm.

Although frequent traversal sequence (FTS) mining has been extensively studied over the last decade in Web usage mining, it is challenging to extend the mining technique to dynamic Web click streams. The main challenge is that existing false-positive methods control memory consumption and output accuracy by a relaxation ratio r (r = e/s, e is the error parameter, and s is the specified minimum support). However, the higher the value of r, the more saving is the memory consumption and the better recall but degrades the output precision, while on the contrary, decreasing r gives a more precise output but needs higher storage space. In this paper, the upper and lower bounds are established to constrain r, a weighted harmonic average (WHA) of the two bounds is designed to adjust r, and a novel algorithm FTS-Stream is proposed to find the FTS over a time-sensitive sliding window. Thus, the precision and recall can be maintained with the WHA (r). Our analysis and experiments show that FTS-Stream has high accuracy and requires less memory in dynamic Web clickstreams

An important application of sequential mining technique is frequent traversal sequence (FTS) mining. However, the Web data grows quickly, some data may be outdated, and previous FTS may be changed when the database is updated. We have to re-mine FTS from the updated database, but re-finding FTS consume too much execution time. In this paper, a novel structure, IE-LATTICE (improved extended lattice) is designed to store the previous FTS. An efficient algorithm based on bidirectional constraint, IMFTS (incremental mining frequent traversal sequence) is proposed, which utilizes the previous mining results and constraint strategy to discover the new FTS just from the added and deleted part of the database. Experimental results show that IMFTS algorithm efficiently reduces the execution time for mining FTS