Paving the Path to Automatic User Task Identification

Web site personalization could be immensely improved if the user's current intentions could be recognized by the surfing behavior. The latter can be captured in the form of events occurring in the browser, like mouse moves or opening Web pages. But which aspects of the user's behavior best contribute to the recognition of the task a user is performing? Is it the number of mouse clicks, the amount of time spent on each page, the use of the back button or anything else? First results of an exploratory study give hint that already simple attributes, such as the average page view duration, the number of page views per minute and the number of different URLs requested, may be usable for the automatic user task identification. 20 participants solved exemplary exercises which corresponded to the user tasks Fact Finding, Information Gathering and Just Browsing. Due to the event logging, true display times were identified, even cached pages and the use of browser tabs were recorded.