2nd ACM Workshop on
Mobile Systems for
Computational Social Science

At UbiComp 2013, Zurich, Switzerland (Sep 9)

The programme has been updated.

For decades, behavioral scientists have struggled to understand the various factors that influence behavior. Observational and self-report methods have shed some light on those factors, but the nature of such methods can elicit responses that are not always completely accurate. Furthermore, such methods usually capture behavior as it occurs in a laboratory as opposed to behavior as it naturally occurs in everyday life. Recent advances in mobile technology provide some extremely powerful tools for overcoming these obstacles.

Indeed, mobile phones represent an ideal platform for studying behavior and interactions in real-life contexts. One reason is because mobile phones are ubiquitous: there are billions of mobile phone users and the market has seen unprecedented growth in recent years. Secondly, mobile phones are unobtrusive: because of their ubiquity, users are not generally aware of the presence of mobile phones, unlike behavioral monitoring in laboratories or through purpose-built devices that depend on self-reports. Thirdly, mobile phones are powerful and sensor rich platforms: modern mobile phones have many sensors embedded in them (e.g., accelerometer, Bluetooth, GPS, and magnetometer) that can accurately capture user behavior; they are also equipped with powerful processors, which allow applications to exploit computationally intensive algorithms to run locally on the phones. There are many open challenges in terms of system design: for instance, since mobile phones are battery powered, efficient algorithms have to be developed to derive accurate inferences from sensor data, and cloud resources can be exploited to support complex computations. Finally, given their diffusion, mobile phones allow to build systems at scale, i.e., supporting mobile applications potentially used by millions of people at the same time. Data collected by means of mobile phones can then be used for analysis of human behavior and interactions. There are also many challenges in this area, especially related to the management of personal data and real-time processing of information. Mobile systems will represent a key foundational component of the emerging discipline of computational social science.