Adaptive User Interfaces and Personalization

In the mid-1990's, I became aware that, in many domains, advisory systems were proving more successful that expert systems, since the domain users wanted support rather than replacement. Clearly, one could use machine learning to acquire the knowledge for such advisory systems, which had led to earlier interest in "learning apprentices". However, the increasing reliance on graphical user interfaces gave the new advisors a different character from the older systems, supporting new methods for collecting trace data about user behavior and new ways to provide advice. My interest in this topic, and the approach that my colleagues and I have taken, were influenced strongly by Jeff Schlimmer's work on what he called self-customizing software. However, I prefer to call them adaptive user interfaces.

In 1996, I joined the DaimlerChrysler Research and Technology Center, where I started three projects along these lines. One involved the Adaptive Route Advisor, which offered a personalized service for in-car navigation based on profiles learned from user decisions. Another system, the Adaptive Place Advisor, carried out a spoken-language dialogue to help users decide on a suitable destinations, such as restaurants, and utilized these conversations to update its user profile. We also collaborated with Daniel Billsus and Michael Pazzani at UCI on their News Dude system, which learned, from implicit feedback, which news stories to read to its users. Another system, INCA, developed in parallel at ISLE, applied similar ideas to an interactive system for crisis planning and scheduling, with user profiles again being based on past traces of user behavior.

Since then, we have utilized this approach to unobtrusive personalization to develop a number of other prototype systems that automatically learning about users from interacting with them. These include an adaptive music player, a personalized travel agent, an adaptive bookmarking system, an online apartment finder, and, most recently, a personalized stock tracker that helps users detect stocks they may want to trade. We have also developed a flexible software environment that makes the construction of such adaptive users interfaces efficient and effective.

Related Publications

Thompson, C. A., Göker, M., & Langley, P. (2004). A personalized system for conversational recommendations. Journal of Artificial Intelligence Research, 21, 393-428.

Yoo, J., Gervasio, M., & Langley, P. (2003). An adaptive stock tracker for personalized trading advice. Proceedings of the International Conference on Intelligent User Interfaces (pp. 197-203). Miami, Florida.

Fiechter, C.-N., & Rogers, S. (2000). Learning subjective functions with wide margins. Proceedings of the Seventeenth International Conference on Machine Learning (pp. 287-294). Stanford, CA.

Rogers, S., Fiechter, C.-N., & Thompson, C. (2000). Adaptive user interfaces for automotive environments. Proceedings of the IEEE Intelligent Vehicles Symposium. Dearborn, MI.

Langley, P., Thompson, C., Elio, R. & Haddadi, A. (1999). An adaptive conversational interface for destination advice. Proceedings of the Third International Workshop on Cooperative Information Agents (pp. 347-364). Uppsala, Sweden.

Gervasio, M. T., Iba, W., & Langley, P. (1999). Learning user evaluation functions for adaptive scheduling assistance. Proceedings of the Sixteenth International Conference on Machine Learning (pp. 152-161). Bled, Slovenia: Morgan Kaufmann.

Langley, P. (1999). User modeling in adaptive interfaces. Proceedings of the Seventh International Conference on User Modeling (pp. 357-370). Banff, Alberta: Springer.

Rogers, S., Fiechter, C., & Langley, P. (1999). An adaptive interactive agent for route advice. Proceedings of the Third International Conference on Autonomous Agents (pp. 198-205). Seattle: ACM Press.

Rogers, S., Fiechter, C., & Langley, P. (1999). A route advice agent that models driver preferences. Proceedings of the AAAI Spring Symposium on Agents with Adjustable Autonomy. Stanford, CA: AAAI Press.

Langley, P., & Fehling, M. (1998). The experimental study of adaptive user interfaces (Technical Report 98-3). Institute for the Study of Learning and Expertise, Palo Alto, CA.

Iba, W., Gervasio, M., Langley, P., & Sage, S. (1998). Evaluating computational assistance for crisis response. Proceedings of the Twentieth Annual Conference of the Cognitive Science Society. Madison, WI: Lawrence Erlbaum.

Gervasio, M., Iba, W., & Langley, P. (2008). Case-based seeding for an interactive crisis response assistant. Proceedings of the AAAI-98 Workshop on Case-Based Reasoning Integrations. Madison, WI.

Gervasio, M., Iba, W., & Langley, P. (1998). Learning to predict user operations for adaptive scheduling. Proceedings of the Fifteenth National Conference on Artificial Intelligence (pp. 721-726). Madison, WI: AAAI Press.

Rogers, S., & Langley, P. (1998). Interactive refinement of route preferences for driving. Proceedings of the AAAI Spring Symposium on Interactive and Mixed-Initiative Decision-Theoretic Systems. Stanford, CA: AAAI Press.

Gervasio, M., Iba, W., Langley, P., & Sage, S. (1998). Interactive adaptation for crisis response. Proceedings of the AIPS-98 Workshop on Interactive and Collaborative Planning (pp. 29-36). Pittsburgh, PA.

Rogers, S., Langley, P., Johnson, B., & Liu, A. (1997). Personalization of the automotive information environment. Proceedings of the Workshop on Machine Learning in the Real World: Methodological Aspects and Implications. Nashville, TN.

Langley, P. (1997). Machine learning for adaptive user interfaces. Proceedings of the 21st German Annual Conference on Artificial Intelligence (pp. 53-62). Freiburg, Germany: Springer.

For more information, send electronic mail to langley@isle.org

© 1997 Institute for the Study of Learning and Expertise. All rights reserved.