Investigations in collaborative multi-party discourse

In this paper, we discuss the efforts underway at the Pacific Northwest National Laboratory in understanding the dynamics of multi-party discourse across a number of communication modalities, such as email, instant messaging traffic and chat data. Only by understanding how individuals communicate through these new media technologies can we hope to successfully design and implement the social media applications of the future. Two prototype systems are discussed: (1) the Conversation Analysis Toolkit (ChAT) is a set of experimental computational linguistic components that enables users to easily identify topics or persons of interest within multi-party conversations, including who talked to whom, when, the entities that were discussed, etc., and (2) the Retrospective Analysis of Communication Events (RACE) application, leveraging many of the ChAT components, is an application built specifically for knowledge workers and focuses on merging different types of communications data so that the underlying message can be discovered in an efficient, timely fashion.