Social Networks

With the arrival of new technology platforms for online interaction and real-time communication enabled by the Internet, as well as the proliferation of more advanced sensors and tracking devices, we now produce vast amounts of data every day, detailing our lives, preferences, friendships, and health. These technologies have not only dramatically changed our lives, but also promise to transform how we study social behavior and dynamics. Yet a complete answer to the question of how individuals make decisions in groups remains elusive. Such studies necessitate the merging and further advancement of both social science and data processing and analysis, by studying interactions, exchanges, and dynamics over large networks of interconnected individuals. IDSS is ideally placed to play a leading role in this endeavor by blending expertise that spans mathematical systems theory, economics, political science, algorithmic and computational game theory, and network science. IDSS research will address such topics as: 1) Developing empirically grounded theoretical frameworks for analysis of information flow, communication, influence, learning, and cascades in social networks, 2) Designing efficient, local, and scalable algorithms for inference with social data, 3) Designing incentive mechanisms for steering behavior towards desired outcomes over large evolving networks, and 4) Developing new architectures for the exchange of information, social interaction, and crowdsourcing.

Read more about IDSS research in social networks in this article: “Using data from social networks to understand and improve systems” (IDSS/MIT News).

Examples of topics in this domain include:

  • How do opinions, social norms, and information evolve over networks?
  • How can we quantify the influence and importance of different individuals situated in networks?
  • How do we assess the impact of social cascades on networks?
  • How can we use social data to make efficient predictions?
  • How do we design recommendation systems that can capture people’s choices efficiently?
  • How can we design better platforms for online markets (e.g., crowdsourcing systems, online matching markets)?

Evolution of Cultural Norms and Dynamics of Sociopolitical Change

Daron Acemoglu, Fotini Christia, Munther Dahleh, Ali Jadbabaie, Asuman Ozdaglar
The recent events in the Arab world have made it clear that questions related to political change, cultural dynamics, and societal transformations are not only of first-order importance for social science, but are also central to developing a scientific approach to policy making and planning. While advances in traditional game theory, political economy, development economics and political science have enabled a posteriori analysis, understanding and predicting sociopolitical change requires a new set of tools and a multidisciplinary analytical framework. This Department of Defense Multi-Investigator University Research Initiative (MURI) project brings together a world-class team of researchers to address this challenge. The project is focused on developing rigorous theory, modeling, and empirical analysis of patterns of communication, interaction, and learning in networked societies.

Consumer Choice-Modeling Software

Devavrat Shah
MIT researchers used novel analytics to refine the process of choosing which products will sell best—revealing insights into how retailers can optimize their shelf space. MIT spinout Celect, co-founded by MIT professors Vivek Farias and Devavrat Shah, develops software that uses a store’s sales and inventory and online buying data to determine which products local customers want to buy.

Identifying Key Individuals in a Research Network

Esther Duflo
This research shows that in India, information diffuses in a social network depending on the position of those first informed. If these injection points are central enough, diffusion takes off; peripheral seeds can stunt diffusion. Researchers demonstrated theoretically and empirically that members of the community are network-central and conducted an experiment demonstrating that seeding information with those who are identified as central by other community members leads to considerably wider information diffusion than seeding information with randomly chosen individuals and village leaders.

Social Networks and Collaborative Problem Solving

Sandy Pentland
Sandy Pentland and his group study how teams of scientists or engineers collaborate on projects using their social networks to gather new ideas and feedback. By providing a link between the literature on team performance and information networks, the authors study teams’ problem-solving abilities as a function of both their within-team networks and their members’ extended networks. As part of this work, they show that while an assigned team’s performance is strongly correlated with its networks of expressive and instrumental ties, only the strongest ties in both networks have an effect on performance.

MIT Institute for Data, Systems, and Society
Massachusetts Institute of Technology
77 Massachusetts Avenue
Cambridge, MA 02139-4307