“Using Big Data, Computational Mechanics and Agent-based Modeling to Understand Social Media Diffusion”

Location: LTS Auditorium, 8080 Greenmead Drive

Speaker:
William Rand
Robert H. Smith School of Business and UMIACS

Abstract:
With the increasing abundance of digital footprints left by human interactions in online environments, the ability to model complex human behavior has become possible. Many approaches have been proposed, but most previous model frameworks are fairly restrictive.

We’ll explore a new social modeling approach that enables the creation of models directly from data with no previous restrictions on the data. We contrast this approach with a variety of other approaches and compare their predictive and descriptive abilities on a heterogeneous catalog of human behavior collected from 15,000 Twitter users.

We ask two questions: can we predict if an individual will tweet, and what is the time when the most number of users are involved in retweeting behavior? Our approach generates individual-level models, which enable us to understand not only the behavior of the aggregate, but also how users affect one another. These models can be compared directly to social theory of how humans behave. Using these models, we can also automatically create an agent-based model that enables the exploration of different policies and interventions on the system. We illustrate examples of how this capability can provide a decision support tool to help policymakers and managers make decisions about communication and targeting.

Bio:
William Rand is an assistant professor of marketing and a member of UMIACS.

He is also the director at the Center for Complexity in Business and holds affiliate appointments with the Departments of Decision, Operations & Information Technology and Computer Science.

Rand’s work examines the use of computational modeling techniques, such as agent-based modeling, geographic information systems, social network analysis, and machine learning to help understand and analyze complex systems.

Over the course of his research, he has used computer models to help understand a large variety of complex systems, such as the evolution of cooperation, suburban sprawl, traffic patterns, financial systems, and more. Rand has received research awards from Google/WPP, the National Science Foundation, DARPA, the Department of Transportation, and the Marketing Science Institute.

He received his doctorate in computer science from the University of Michigan in 2005.