Tina Eliassi-Rad  |  Associate Professor of Computer Science  |  Rutgers University

Thursday, March 10, 2016 |  7pm  |  Goertzen Communication Addition 21

Distinguished Speaker Series in Data Science- Learning, Mining and Graphs

In this talk, I will discuss three dependent graph-mining problems ranging from theory to algorithms to applications. (1) Measuring tie-strength: Given a set of people and a set of events attended by them, how should we measure connectedness or tie strength between each pair of persons? The underlying assumption is that attendance at mutual events produces an implicit social network between people. (2) Role discovery: given a graph, how can we automatically discover roles (or functions) of nodes? Roles should compactly represent structural behaviors of nodes and generalize across various graphs. (3) Network similarity: Given two networks (without known node-correspondences), how should we measure similarity between them? This problem occurs frequently in many real-world applications such as efficacy of transfer learning, re-identification, and change detection.