Technology has evolved a lot since then, but closets have been largely untouched by innovation. Now, that's starting to change. "If algorithms do their job well, people will spend less time thinking about what to wear," said Ranjitha Kumar, an assistant professor in the Department of Computer Scie...

The Seventh International Conference on Intelligent User Interfaces (IUI-2003) was held 12-15 January 2003 in Miami Beach, Florida. The conference brought together researchers and practitioners to report on outstanding research and applications, examine emerging work, and delineate new avenues for intelligent user interfaces. The conference received an all-time record number of submissions, covering a wide range of areas and approaches. Oral presentation sessions were organized into seven major topics: (1) adaptive and collaborative interfaces, (2) affective interfaces, (3) agent-based interfaces, (4) knowledge acquisition and visualization, (5) model-based interface design, (6) multimodal input, and (7) natural language interfaces. The conference program included three invited talks, each reflecting a different direction for developing the intelligent interfaces of the future.

Research on mixed-initiative interaction and assistance is still in its infancy but is poised to blossom into a wellspring of innovation that promise to change the way we work with computing systems--and the way that computing systems work with us. I share reflections about the opportunities ahead for developing computational systems with the ability to engage people in a deeply collaborative manner, founded on their ability to support fluid mixed-initiative problem solving. Such collaborative intelligence sits at the veritable heart of human civilization. In the course of daily life, we assume and rely on a rich interleaving of efforts to achieve goals while immersed in shared context. We continue to engage one another in efficient, tightly woven collaborations, reasoning with remarkable efficiency about the beliefs, preferences, intentions, and skills of potential collaborators. The inferences underlying successful collaborations typically stream in such an effortless and subconscious manner that we often fail to recognize the elegance and sophistication of these capabilities. The magic of human collaborative competency comes to the foreground with attempts to extend these skills to computational systems. Developing a better understanding of the core aspects of intelligence that enable people to collaborate with fluidity promises to enable new kinds of human-computer collaboration. The nascent area of research on mixed-initiative interaction centers on developing methods that enable computing systems to support an efficient, natural interleaving of contributions by people and computers, aimed at converging on solutions to problems. In mixed-initiative interaction, people and computers take initiatives to contribute to solving a problem, achieving a goal, or coming to a joint understanding. Conversational dialogue is an oft-cited example of mixed-initiative interaction, referring to the ability of each participant in a dialogue to take initiative to guide or add to a discussion. Endowing an automated dialogue system with the ability both to take initiative ("What city do you wish a flight to?") and to allow people to take conversational initiative ("Wait, I'd like to add a side trip.") However, mixed-initiative interaction extends beyond spoken conversations to include a broad spectrum of collaborative problem solving marked by an interleaving of contributions by different participants. Mastering mixed-initiative interaction poses a constellation of fascinating challenges and opportunities for AI researchers. Figure 1 highlights the core challenge of seeking mutual understanding or grounding of joint activity. Joint activity describes the behavior displayed by people working together to solve a mutual goal.

Interface personalization aims to streamline the process of working in a feature-rich application by providing the user with an adapted interface tailored specifically to his or her needs. The mixed-initiative customization assistance (MICA) system explores a middle ground between two opposing approaches to personalization: (1) an adaptable approach, where personalization is fully user controlled, and (2) and adaptive approach, where personalization is fully system controlled. We overview MICA's strategy for providing user-adaptive recommendations to help users decide how to personalize their interfaces. In doing so, we focus primarily on how MICA handles threats to usability that are often found in adaptive interfaces including obtrusiveness and lack of understandability and control. We also describe how we evaluated MICA and highlight results from these evaluations.

The field of human-computer interaction (HCI) offers designers and developers of interactive systems a large repertoire of methods for ensuring that their systems will be both usable and useful. This article offers a brief introduction to these methods, focusing on the ways in which they may need to be adapted and extended to take into account the characteristic properties of systems that include some sort of AI. The discussion is organized around three types of activity: understanding users' needs, interaction design, and evaluation. The application of these methods is described with terms such as usability engineering and user-centered design. These activities constitute part of the overall software development process, and they must be carefully coordinated with the other activities involved in that process.

Various techniques were presented for eliciting the decision model incrementally in conjunction with the problem-solving process. Well-established techniques from decision analysis, including sensitivity analysis and value of information calculation, were also discussed in the context of incremental model elicitation. Finally, the importance of selfexplanatory systems was emphasized because the user needs to understand the impact of his/her communicated preferences and their role in the problem-solving process. The American Association for Artificial Intelligence Spring Symposium on Interactive and Mixed-Initiative Decision-Theoretic Systems was held at Stanford University from 23-25 March 1998. The symposium attracted approximately 30 researchers from around the world.

Adaptive graphical user interfaces (GUIs) have the potential to improve performance and user satisfaction by automatically tailoring the presentation of functionality to each individual user. In practice, however, many challenges exist, and evaluation results of adaptive GUIs have been mixed. To guide researchers and designers in developing effective adaptive GUIs, we outline a design space and discuss three important aspects to consider when conducting user evaluations of these types of interfaces: the control and reporting of adaptive algorithm characteristics, the impact of task choice and user characteristics on the overall effectiveness of a design, and evaluation measures that are appropriate for adaptive interaction. A familiar example of an adaptive interface is the Windows XP start menu, where a small set of applications from the "All Programs" submenu is replicated in the top level of the "Start" menu for easier access, saving users from navigating through multiple levels of the menu hierarchy (figure 1). The potential of adaptive interfaces to reduce visual search time, cognitive load, and motor movement is appealing, and when the adaptation is successful an adaptive interface can be faster and preferred in comparison to a nonadaptive counterpart (for example, Gajos et al. [2006], Greenberg and Witten [1985]).

John Deere Dubuque Works (Dubuque, Iowa), a manufacturer of agriculture and industrial equipment, has implemented a virtual-reality system to use in its construction division. The system enables John Deere to use virtual-product prototypes to assess key design factors in construction equipment, such as visibility and the ability to reach controls. Nabisco Biscuit (East Hanover, N.J.), a manufacturer of cookies and crackers, has installed an intelligent process-operating guidelines (POG) system. This POG system uses expert system technology to provide realtime process control information to the bakeries. Automotive manufacturer Ford Powertrain Operations (Dearborn, Mich.) has developed a flexible manufacturing system that is being controlled by an intelligent cell controller.

Similar systems are being installed at other Texaco sites. Lear Astronics (Santa Monica and Ontario, Calif.) is combining neural networks with virtual reality to enhance its Autonomous Landing Guidance (ALG) system. Lear Astronics is using a neural network-based massively parallel coprocessor for real-time image processing in the ALG system, which enables commercial and military aircraft pilots to land in foggy conditions. Researchers at Georgia Tech (Atlanta, Ga.) have created intelligent agent software called the Technology Opportunities Analysis Knowbot (TOAK) that provides profiles of the latest technological trends and opportunities. TOAK navigates through multiple networks and across diverse computer systems to perform specific search tasks for the user.

BNR Europe (Harlow, England), the R&D subsidiary of telecommunications equipment supplier Northern Telecom, is using virtual reality technology for equipment installation planning. The VR system allows BNR's engineers to visualize complex installations and how they will work, greatly saving time and effort compared to the traditional CAD system. Continental Bank (Chicago, Ill.) has developed a client/server-based intelligent application to improve the quality of its customer service. Thanks to an expert system, the bank's service management staff has immediate access to customers' cash management account information online. Anderson Memorial Hospital (Anderson, S.C.) had implemented a neural network-based hospital information and patient prediction system which has improved the quality of care, reduced the death rate and saved the facility millions of dollars in resources.