Inference in Complex Social Systems: Insights and Applications from the Behavior of the Aggregate

I have used mobile phones to continuously gather information including proximity, location, and communication from 100 human subjects at MIT. Systematic measurements from these people over the course of nine months has generated one of the largest dataset of continuous human behavior ever collected, representing over 300,000 hours of daily activity. Additionally, in collaboration with one of Europe's major telecommunication companies, I am currently analyzing the call logs of an entire country - a dynamic social network consisting of 250 million nodes and 12 billion temporal edges.

 In this talk I describe how this type of data can be used to uncover the structure in behavior of both individuals and organizations, infer relationships, and study social network dynamics. By combining theoretical models with rich and systematic measurements, we show it is possible to gain insight into the underlying behavior of complex social systems.

 While results such as uncovering scaling laws from the communication patterns of hundreds of millions of people will certainly be one emphasis in this talk, of equal importance is how this data can enable applications that improve our society. I will demonstrate a variety ways these insights into our own behaviors can be used to develop applications that better support both the individual, organization and society.

 Finally, I will overview my current Nokia-funded project, EPROM (Entrepreneurial Programming and Research on Mobiles), an initiative to develop a mobile phone programming curriculum currently being deployed across Africa. There are over 100 undergraduate computer science students from the University of Nairobi in Kenya and the GSTIT in Ethiopia actively participating in the program and prototyping their own mobile phone applications in an area that is currently the fastest growing mobile phone market in the world.

 More information about these projects is available here:

http://reality.media.mit.edu

http://mit.edu/eprom