Mining browsing and purchasing behaviors of web users

Mining association rules is to discover associations among purchased items, which are valuable for cross-marketing, catalog design, add-on sales, store layout and customer segmentation. Mining web traversal patterns is to discover traversal sequences for most of the web users, which can provide navigation suggestions for web users. However, association rules cannot describe navigation behaviors of web users, and web traversal patterns cannot describe purchasing behaviors of web users. To overcome these disadvantages, this paper proposes a method to simultaneously discover both navigation and purchasing behaviors of customers, such that the requirements of the web site managers can be satisfied.