Relevance of Time Spent on Web Pages
Peter I. Hofgesang · 2006
What is the intention of an online visitor? This is one of the most valuable pieces of information for an online service provider. In the case of a real-world shop, customers have the ability to explicitly express what they are looking for. However, on the web, the intention is hidden and can only be partially revealed from implicit indicators in the traces users leave behind while they browse through a website. The vast majority of researchers in web usage mining exploit only two types of information: the order of visited pages and their popularity- i.e. the number of times they were visited. However, several studies in information retrieval and human-computer interaction have suggested a third factor, the time spent on web pages (TSP), as an important measure of intention and relevance. The key contributions of this paper are: (1) an extensive survey of possible factors that influence the TSP measure, and (2) a similarity measure that applies to TSP and can be used to cluster users based on their assumed intentions. Our experiments are based on log files generated by several commercial websites. Key Words: web usage mining, time spent on web pages, clustering 1.