MIE Distinguished Seminar Series: Susan Dumais, Microsoft | The Potential for Personalization in Web Search

October 26, 2018

Mechanical Engineering Building, MC102

The Potential for Personalization in Web Search

Traditionally web search engines returned the same results to everyone who asks the same question.  However, using a single ranking for everyone in every context at every point in time limits how well a search engine can do in providing relevant information.  In this talk I present a framework to quantify the “potential for personalization” which is used to characterize the extent to which different people have different intents for the same query.  I describe several examples of how different types of contextual features are represented and used to improve search quality for individuals and groups.  Finally, I conclude by highlighting important challenges in developing personalized systems at Web scale including privacy, transparency, serendipity, and evaluation.


Susab Dumais is a technical fellow and managing deputy director of MSR AI. Her research is at the intersection of information retrieval and human-computer interaction. She is interested in algorithms and interfaces for improved information retrieval, as well as general issues in human-computer interaction.

She has been at Microsoft Research since July 1997. Her current research focuses on gaze-enhanced interaction, the temporal dynamics of information systems, user modeling and personalization, novel interfaces for interactive retrieval, and search evaluation. Previous research studied a variety of information access and management challenges, including personal information management, desktop search, question answering, text categorization, collaborative filtering, interfaces for improving search and navigation, and user/task modeling. She has worked closely with several Microsoft groups (Bing, Windows Desktop Search, SharePoint Portal Server and Office Online Help) on search-related innovations.

Prior to coming to Microsoft, Susan co-developed a statistical method for concept-based retrieval known as Latent Semantic Indexing. You can find pointers to this work on the Bellcore (now Telcordia) LSI page.