Date of Award

Spring 5-15-2022

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Electrical Engineering and Computer Science


Soundarajan, Sucheta


Assignment, Fairness, Information Flow, Social Networks

Subject Categories

Computer Sciences | Physical Sciences and Mathematics


In professional and other social settings, networks play an important role in people's lives. The communication between individuals and their positions in the network, may have a large impact on many aspects of their lives.In this work, I evaluate fairness from different perspectives.First,tomeasurefairnessfromgroupperspective,Iproposethenovelinformation unfairness criterion, which measures whether information spreads fairly to different groups in a network. Using this criterion, I perform a case study and measure fairness in information flow in different computer science co-authorship networks with respect to gender. Then, I consider two applications and show how to increase fairness with respect to a fairness metric. The first application is increasing fairness in information flow by adding a set of edges. I propose two algorithms- MaxFair and MinIUF- which are based on detecting those pairs of nodes whose connection would increase flow to disadvantaged groups. The second application is increasing fairness in organizational networks through employee hiring and assignment. I propose FairEA, a novel algorithm that allows organizations to gauge their success in achieving a diverse network. Next,Iexaminefairnessfromanindividualperspective.Iproposestratification assortativity, a novel metric that evaluates the tendency of the network to be divided into ordered classes. Then, I perform a case study on several co-authorship networks and examine the evolution of these networks over time and show that networks evolve into a highly stratified state. Finally, I introduce an agent-based model for network evolution to explain why social stratification emerges in a network.


Open Access