Balancing the Needs of Personalization and Reasoning in a User-centric Scheduling Assistant

We describe the interaction of three aspects core to a personalized scheduling task. First, we develop a preference model designed to capture user preferences for the task of scheduling a meeting request between multiple people, and a methodology for preference elicitation to initially populate this model. Second, we explain a natural-language-based elicitation of the meeting request details and constraints, and outline the solving of the resulting constrained scheduling problem (with preferences). Third, we describe the display of solutions to the scheduling problem to the user, as candidate scheduling options with explanations, and detail unobtrusive learning of revisions to the preference model from the user's choices among the candidates. We describe the user studies that informed our design choices, and assess the resulting system in terms of the quality of scheduling options presented, according to the user. The scheduling task enabled by the integration of these aspects has been implemented within a deployed application.