Online activity graph for document importance and association

Abstract

The way in which a user interacts with her desktop while performing some task generates an information trail that can be used to identify the task context and the user's interests. This new information can in turn be fed back into the system to increase the level of support available to the user for both current and future tasks. In this paper we present research which analyses user-activity log les to explore how a user's activities evolve with time. Resources fall in and out of a task based on the user's mental model for tackling that task. We assign time-varying, importance and association values to each resource, based on the dwell-time and the resource-switching patterns exhibited by the user while browsing. Furthermore, we propose a new dynamic graph algorithm called OnlineActivityGraph which leverages on these values to generate document clusters and short term user models. We further present a discussion about the encouraging results obtained from our preliminary experiments.