Site Browsing and Purchase Behavior in an EC Site

Abstract

Electronic Commerce (EC) is grown as an important purchase channel. Each EC company make some strategies to obtain and maintain his customer. Individual recommendation is one of the representative strategies. In this paper, we propose a quantitative model to discover signs of purchase, and consider the effectiveness of our model through a real-data analyzing. As our example, we analyze the browsing behavior of a golf goods EC site using the web access log and purchase record data. First, to divide into some homogeneous groups of customers, we utilize some characteristic variables to explain the customers’ behavior, then we do cluster analysis. Next we evaluate the purchase behavior for each group using logistic regression analysis.