Seminar on Computational Learning and Adaptation

  Adaptive User Interfaces for Personalized Services

Pat Langley

Institute for the Study of Learning and Expertise
Palo Alto, California
and
Center for the Study of Language and Information
Stanford University

http://cll.stanford.edu/~langley

The Internet has made available increasing amounts of information and given users more choices than ever before, but all too often the result is more confusion than satisfaction. To some extent, recommendation systems can help people filter relevant information and guide their choices, but users have different goals and distinctive tastes. In this talk, I review the results of our research programme on adaptive user interfaces - interactive systems that automatically personalize their content to individual users. Adaptive interfaces incorporate technology and principles from machine learning, intelligent agents, and human-computer interaction to improve the user's experience. Over the past five years, we have designed and implemented many prototype systems of this sort, including a navigation aide, destination advisor, news reader, music player, travel agent, apartment finder, interactive scheduler, and stock tracker. I consider key decisions that arose in the design of these systems, present some evidence that they adapt effectively to individual users, and propose a general framework for characterizing such personalized assistants. In closing, I claim that progress in adaptive interfaces depends not on the development of new algorithms, but rather on the integration of known methods in new ways.


This talk describes joint work with Daniel Billsus, Nicolas Fiechter, Melinda Gervasio, Wayne Iba, Mehmet Goker, Mike Pazzani, Seth Rogers, Cynthia Thompson, and Jungsoon Yoo.

Return to the seminar schedule