Using pattern-join and purchase-combination for mining Web transaction patterns in an electronic commerce environment

Explores a data mining capability which involves mining Web transaction patterns in an electronic commerce (EC) environment. To better reflect the customer usage patterns in the EC environment, we propose a data mining model that takes both the customers' travelling patterns and their purchasing behaviour into consideration. We devise two efficient algorithms [MTS/sub PJ/ (Maximal Transaction Segment with Pattern Join) and MTS/sub PC/ (Maximal Transaction Segment with Purchase Combination)] for determining frequent transaction patterns, which are termed "large transaction patterns" in this paper. In addition, the WTM (Web Transaction Mining) algorithm is used for comparison purposes. By utilizing the path-trimming technique, which is developed to exploit the relationship between travelling and purchasing behaviours, MTS/sub PJ/ and MTS/sub PC/ are able to generate the large transaction patterns very efficiently. A simulation model for the EC environment is developed and a synthetic workload is generated for performance studies.