6 Jul 2015
Speaker: Prof Padhraic Smyth, Department of Computer Science, University of California, Irvine
Title: Latent Variable and Temporal Event Models for Network Data
Date: Monday 6 July 2015
Location: Room 4.31/4.33, Informatics Forum, University of Edinburgh
Chair: Chris Williams, Director of the CDT in Data Science, University of Edinburgh
Social network analysis has a long history in the social sciences, often with a focus on relatively small survey-based data sets. In the past decade, driven by the ease of collecting network information in an automated manner (e.g., for social media and phone networks) there has been significant interest in developing machine learning techniques for network data. This has led to an increased emphasis on topics that were traditionally beyond the scope of traditional social network analysis: integration of non-network data (such as text), scalability to large networks, and predictive evaluation. In this talk we will discuss recent progress on two classes of models in this general context: latent variable models for static networks and relational event models for temporal networks. We will review the representational capabilities of these models from a generative perspective, discuss some of the challenges of parameter estimation that arise in this context, and emphasize the role of predictive evaluation for network modeling. The talk will conclude with a brief discussion of future directions in this general area.
Based on joint work with Chris DuBois, Carter Butts, and Jimmy Foulds.