Resnik Developing Computational Models to Better Understand How Decisions Are Made
![](https://www.umiacs.umd.edu/sites/default/files/styles/medium/public/resnik%20NSF%20grant.jpg?itok=_zCS40rR)
A University of Maryland expert in computational linguistics has received funding from the National Science Foundation (NSF) to develop computational methods to advance our understanding of how decisions are made, particularly in the context of the political arena.
Philip Resnik, a professor of linguistics with a joint appointment in the University of Maryland Institute for Advanced Computer Studies (UMIACS), is principal investigator of the $430K award, which comes from the NSF’s Division of Information and Intelligent Systems.
The two-year project will examine both metadata about individuals and the language they use as sources of information that can help predict when and why two individuals will make the same choice.
“This is a great opportunity not only to develop new computational techniques, but to bridge the gap between technology and social science,” says Resnik.
In many settings, Resnik says, groups are often presented with a specific decision to act upon. Legislators, for example, can be presented with a bill to vote on; a set of scientific authors can decide whether or not to cite a piece of research in their papers; or social media users might decide whether or not to share specific content they see online.
The NSF project will use advanced natural language processing and machine learning techniques to develop a much deeper understanding of what goes into these types of decision-making processes, Resnik says, starting with legislative decisions and then expanding beyond the political domain.
The researchers plan to go beyond previous computational models used in political science, which often focus on the properties of the individuals—for example, whether they come from the same political party, or have demographic characteristics in common.
What the researchers believe will bring added value to their analyses and comparisons is detailed statistical modeling of the language that legislators use when talking about issues.
If legislators frame issues in similar ways, for example, does that mean they are more likely to make similar decisions where those issues are concerned? Does the use of emotional language magnify persuasive power, or, conversely, might it instead serve to increase polarization?
“Colleagues are an important influence on legislators’ decisions, but these dynamics are hard to observe,” says Kris Miler, an associate professor in the Department of Government and Politics who is co-PI on the project. “By examining shared language, this research will provide new insight into how legislators work together, how policy ideas can gain momentum, and how legislative relationships are important even in these partisan times.”
Central to the project are Alexander Hoyle, a second-year doctoral student in computer science, and Pranav Goel, a third-year doctoral student in computer science.
Resnik, Hoyle and Goel are all active in the Computational Linguistics and Information Processing Lab, where the robust computational resources provided by UMIACS are available to crunch the large amounts of data needed for the project.
—Story by Melissa Brachfeld