TBD
Reference Information:
Title: An adaptive calendar assistant using pattern mining for user preference modelling
Author: Alfred Krzywicki, University of New South Wales, Sydney, Australia; Wayne Wobcke, University of New South Wales, Sydney, Australia; Anna Wong, University of New South Wales, Sydney, Australia
Presentation Venue: IUI '10 Proceedings of the 15th international conference on Intelligent user interfaces
Summary:
The researchers are working on a system they call SmartCal. This system is a calendar assistant that uses pattern mining to determine the user's preferences when setting calendar events and appointments. The system will then modify it's presentation such that the user will be displayed the most commonly used input controls when adding or editing in a way that matches their historical and current preferences.
Users have reported to the researchers that they found that appointment creation become more efficient and intuitive for them. In addition to this finding, the users reported that they liked this particular interface more than other appointment making interfaces they had previously used.
Discussion:
When I first started reading about this, I was reminded of the HeliCalendar project that had been proposed in our Senior Capstone Design class. However, that interface was static in every proposal I recall hearing about and this one sounds like it would modify its presentation layer to match your preferences over time.
I like this approach to the design. Since everyone has different preferences, it is often difficult as a developer to predict in advance what the average response will be and what most people will prefer. It is often costly to do user studies as well as being time consuming. If more software developers could learn to build algorithms similar to these into their products, those sorts of studies might become a historical footnote, as the applications themselves will reconfigure around the users desires and be many things to many people.
No comments:
Post a Comment