Project NEEM: Agent Based Meeting Augmentation

This document introduces project Neem, its goals and directions, and its motivation. Project Neem at the University of Colorado is concerned with improving distributed multimedia meetings. A major novel aspect of this research is the use of intelligent artificial agents as full-fledged meeting participants. Goals of this interdisciplinary project include the enhancement of group interaction understanding, and the creation and testing of a prototype distributed meeting environment. Research methods include theoretical modeling, meetings analyses, prototype implementation and testing in real world environments, and meeting metrics development and application. Tools for understanding the social and organizational context of these meetings include the SYMLOG methodology applied by social agents, and the IBIS methodology applied by organizational agents. A new model enabling these agent analyses is the GraspIce model, explained in this document. The work described herein is part of an ongoing effort within the Collaboration Technology Research Group at the University of Colorado to understand human collaboration, and to assist and enhance that collaboration via technology.