Inferring calendar event attendance

Abstract

The digital personal calendar has long been established as an effective tool for supporting workgroup coordination. For the new class of ubiquitous computing applications, however, the calendar can also be seen as a sensor, providing both location and availability information to these applications. In most cases, however, the calendar represents a sequence of events that people could (or should) attend, not their actual daily activities. To assist in the accurate determination of user whereabouts and availability, we present Ambush, a calendar system extension that uses a Bayesian model to predict the likelihood of one's attendance at the events listed on his or her schedule. We also present several techniques for the visual display of these likelihoods in a manner intended to be quickly interpreted by users examining the calendar.