III: Small: Algorithms and Practical Applications for Team Formation and Change

Trustees Of Boston University, Boston

Investigators

Abstract

For decades, researchers in sociology, psychology, organizational behavior, and other social sciences have been conducting quantitative data-analysis studies in order to understand and model how individuals form teams and how different factors (like diversity or homogeneity) affect team performance. Recently, the proliferation of online platforms for education, outsourcing, collaborative learning, software development and problem solving has also motivated computer scientists to work on the development of algorithms for team formation. This work closes the loop between the theory and practice of team-formation research. This is done by integrating social-science findings in the development of new computational problems and the design of new algorithms for team engineering, which is a term used to encompass the formation of teams and the modification of existing teams. Moreover, this work develops the necessary computational concepts as well as the software that enables the application of the team-recommendation algorithms to a variety of educational scenarios including hackathons, student entrepreneurship teams, as well as team formation in curricular and extracurricular project-based learning opportunities. The research will have strong algorithmic, experimental and engineering components and will consist of the following research thrusts. Motivated by research in social sciences, in the first thrust the project will develop computational concepts enabling the formation of teams with low faultline potential. In the second thrust the project will focus on algorithms that modify existing teams in order to improve their functionality; in this thrust concepts that exist in social science research will again be used, viewed through computational lenses. In the final thrust the project will also consider the online team engineering problem where teams are formed on demand and consist both of core members and non-members that are temporarily hired in order to help with a particular task; in this scenario, the focus will be on algorithmic questions related to cost-optimizing strategies. In addition to these problems, the project will also consider problems arising in educational settings (e.g., hackathons, student entrepreneurship teams) and the development of a software platform to implement team-formation algorithms in such practical settings. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.