Designing Efficient and Accurate Behavior-Aware Mobile Systems

15 Aug
Friday, 08/15/2014 6:00am to 8:00am
Ph.D. Thesis Defense

Abhinav Parate

Computer Science Building, Room 142

The proliferation of sensors on smartphones, tablets and wearables has led to a plethora of behavior classification algorithms designed to sense various aspects of individual user's behavior such as daily habits, activity, physiology, mobility, sleep, emotional and social contexts. This ability to sense and understand behaviors of mobile users will drive the next generation of mobile applications providing services based on the users' behavioral patterns. In this thesis, we investigate ways in which we can enhance and utilize the behavioral understanding of users in such mobile applications. In particular, we focus on identifying the key challenges in the following three aspects of behavior-aware applications: detection, analysis and prediction of user behaviors; and present systems and techniques developed to address these challenges.

In this thesis, we first demonstrate the utility of wrist-worn devices equipped with inertial sensors in real-time detection of health-related behaviors such as smoking and eating with high accuracy, recall and precision. Our approach detects these behaviors in a passive manner without any explicit user interaction and does not require use of any cumbersome device. Next, we design a context-query engine to support behavior analytics applications on mobile devices. Our proposed context-query engine performs information fusion across several user contexts for an individual user to enable optimizations like i) energy-efficient continuous context-sensing, and ii) accurate context inference. Finally, we present a behavior prediction algorithm that predicts the phone user's app usage behavior; and utilizes these predictions in improving the freshness of mobile applications such as Facebook that present users with the latest content fetched from the remote servers. We show that our proposed algorithm delivers application content to the user that is on an average fresh within 3 minutes. An implementation of our algorithm is available as a widget in Google Play Store that shows shortcuts for the predicted apps; and has been downloaded and installed on more than 50,000 devices.

Advisor: Deepak Ganesan