We present a method for entity resolution that infers relationships between observed identities and uses those relationships to aid in mapping identities to underlying entities. We also introduce the idea of using graphlets for entity resolution. Graphlets are collections of small graphs that can be used to characterize the “role” of a node in a graph. The idea is that graphlets can provide a richer set of features to characterize identities. We validate our method on standard author datasets, and we further evaluate our method using data collected from Twitter. We find that inferred relationships and graphlets are useful for entity resolution.

We present a method for finding groups of interest in large social networks. A temporal group detection algorithm identifies tightly connected groups by analyzing communications as they unfold over time. Since the number of groups found through temporal group detection may be too large to allow for manual analysis of their behavior, we also present an algorithm to identify groups of interest within an existing set of groups. This algorithm works by observing how groups react to key events and finds groups of interest by noting which groups respond anomalously. We demonstrate this approach on two social media datasets collected from Twitter. The first dataset involves tweets from Afghanistan when Afghanistan signed a letter of cooperation with India. The second dataset involves tweets surrounding the death of Steve Jobs. In both cases, our algorithm was able to identify appropriate groups.