March 30, 2010 - Tanzeem Choudhury

Title: People-Aware Computing: Towards Societal Scale Sensing using Mobile Phones

Abstract:
A great variety of sensors are being built into mobile phones today. This opens up a new research frontier, where mobile devices have the potential to significantly impact many aspects of our everyday life: from health-care, to sustainability, safety, entertainment, and business. However, the broad impact of this vision will be jeopardized without advances in the computational models, which turn raw sensor data into inferences (ranging from recognizing physical activities to tracking community-wide social interaction patterns). My group focuses on developing mobile sensing and machine learning techniques for analyzing and interpreting the behavior of individuals and social groups, including their context, activities, and social networks.

Although solutions to this problem using standard machine learning techniques is possible, how they can be solved efficiently without requiring significant effort from the end-users is still an open problem. In this talk I will describe the research we have done to bridge this gap and advance the state of the art of people-aware computing, by developing novel learning algorithms and evaluating resulting systems through a series of real-world deployments. I will conclude by providing some specific examples of how the methods we have developed can be applied to better understand and enhance the lives of people.