The Correlation Clustering Problem

In correlation clustering, the goal is to cluster data given similarity/dissimilarity judgment about pairs of data items. A variety of associated optimization problems have been studied from the approximation algorithms viewpoint. In this talk, I will survey some of these results. We will encounter linear programming, semi-definite programming, and combinatorial techniques that have been used to attack the problem and take a random walk through several different variants of the basic problem that have been studied over the years.