Predicting the Consumers Behavior with Using the Variety of User Access Patterns

Abstract

It is important for the companies to get much information about their customers. For example, the information that how often each customer purchases from the competitors can be useful when we plan the promotion strategy. In this paper, we focus on the web services and predict customers' behaviors such as browses and purchases in two companies and their competitors. We define "the variety" of User Access Patterns which has not received much attention in previous studies and show its usability to predict customers behaviors.