Prof. Boleslaw Szymanski presented an invited talk on "Algorithms for Robust Community Detection" at the University of Houston Department of Physics, March 3, 2017

The talk describes two robust algorithms for community detection. The first targets communities arising in biological functions carried out by groups of interacting molecules, cells or tissues. These communities may overlap when biological components are involved in multiple functions. Moreover, they may be unstable because traditional methods are sensitive to noise and parameter settings. To address these challenges, we introduce an unorthodox clustering method called SpeakEasy, which identifies communities using top-down and bottom-up approaches simultaneously. The second algorithm is based on modularity optimization, but avoids the well-known resolution limit problem of this approach. We use novel regression model which assigns weights to the edges of a graph according to their local topological features to improve detected communities quality.