An Overview of Microsoft Research

I'll give an informal overview of research efforts at Microsoft Research, focusing on the Adaptive Systems and Interaction Group. Research in this group attempts to enhance the user experience through the synthesis of innovations in machine intelligence and human-computer interaction. Key projects include work on information management, perception and memory, new input and visualization tecnhiques, mixed-initiative interaction, and sensory fusion and learning. My own research is in information retrieval, so many examples will come from
that realm.

Bringing Order to the Web ... and Beyond

In this talk I will describe algorithmic and interface innovations developed to help users organize information, in particular web searchresults. Today web searches return a ranked list of best-matching pagesand users have to sift through these long undifferentiated lists to find items of interest. An important mechanism for facilitating information access in a wide variety of applications is a structured knowledge hierarchy, such as those used by library classification systems and more recently web directories like Yahoo! and LookSmart. Structures like these can be used to automatically organize arbitrary web search results as well. Using our system, a query about "saturn" will pages organized into those having to do with automobiles, computer games, the planet, and a WWF wrestler. There are two key technologies for accomplishing this: 1) developing models for hierarchically classifying arbitrary text pages on-the-fly, and 2) building user interfaces for taking advantage of the resulting structure. I will begin by talking about enhancements to support vector machine (SVM) learning algorithms for text classification that exploit hierarchical structure, and then describe how we developed and evaluated a series of user interfaces to support structured search. Interfaces that show results in context provide retrieval advantages of up to 50% over traditional lists and arestrongly preferred by users.
    Susan Dumais is a Senior Researcher in the Adaptive Systems and Interaction Group at Microsoft Research where she works on algorithms and interfaces for improved information access and management. Prior tojoining Microsoft Research in July 1997, she was at Bellcore and Bell Labs for many years. She received a B.A. in Mathematics and Psychology from Bates College and a PhD in Cognitive Psychology from Indiana University. She has published widely in the areas of human-computer interaction and information retrieval. Her current research focuses on text categorization using inductive learning techniques, and collaborative information retrieval (a joint project with Univ Washington, Boeing and RISO Labs). Previous research included well-known work on Latent Semantic Indexing (a statistical method for concept-based retrieval), combining search and navigation, individual differences, perceptual learning and attention, and organizational impacts of new technology.
    Susan is Chair of ACM's SIGIR group, and serves on the NRC Committee on Computing and Communications Research to Enable Better Use of Information Technology in "Digital Government". She serves on the editorial board of: ACM: Transactions on Information Systems, ACM: Transactions on Human Computer Interaction, Human Computer Interaction, Information Processing and Management, Information Retrieval, Hypertext, Encyclopedia of Information Retrieval, and Annual Review of Information Science and Technology, and is actively incolved on program committeesfor several conferences. She has been a visiting faculty member at Stevens Institute of Technology, New York University, and the Universityof Chicago.