Modeling interactions from email communications

Email plays an important role as a medium for the spread of inf rmation, ideas, and influence among its users. We present a framework to learn topic-based interactions be tween pairs of email users, i.e., the extent to which the email topic dynamics of one user are likely to be affected by the others. The proposed framework is built on the influence model and the probabilistic latent semantic an alysis (PLSA) language model. This paper makes two contributions. First, we model interactions between em ail users using the semantic content of email body, instead of email header. Second, our framework models not on ly email topic dynamics of individual email users, but also the interactions within a group of individua ls. Experiments on the Enron email corpus show some interesting results that are potentially useful to dis cover the hierarchy of the Enron organization. We also present an email visualization and retrieval system which c ould not only search for relevant emails, but also for the relevant email users.