In this talk I will introduce a framework for modeling social network interaction data using Bayesian Networks. I will then show how to incorporate additional information about people, such as their interests and affiliations. Incorporating such data has been shown to improve models based solely on interactions (Wasserman,1994;Hoff et al, 2002). I will describe a generative model that combines block modeling approaches with Bayes Net structure learning learning from both --- interactions and personal information. I will describe a scoring metric that optimizes Bayes Net structure and latent clustering of people iteratively allowing to obtain better quality models. Experimental results show improvement on several fronts: the Bayesian Networks overfit less when additional information is used, and learned latent clusterings of people help to provide meaningful insights into social groupings. This work was done jointly with Zoubin Ghahramani and is based on my thesis (Goldenberg, 2007).