Mission
The IntelliMedia Center for Intelligent Systems is a multidisciplinary laboratory focusing on intelligent human-computer interaction and communication. The fundamental question we explore is, “How can we create flexible, adaptive systems that significantly enhance human problem-solving?” For example, how can we devise educational software that dynamically adapts to individual students? How can we build natural language systems that effectively communicate with their users? By fusing adaptive reasoning, human language technologies, and rich media interfaces, the IntelliMedia team investigates flexible systems that dynamically adapt themselves to their users and their environments. The lab conducts basic research on natural language processing, human-computer interaction, and intelligent user interfaces. Because our work is guided by the belief that progress is accelerated by a tight coupling of theory and practice, we are keenly interested in applications that deal with the complexities of real-world problems, particularly in educational software.

Research
Our research focuses on computational models of intelligent interaction and communication with an emphasis on natural language processing and intelligent user interfaces. We are especially interested in mixed-initiative systems that facilitate human problem-solving and learning by providing context-sensitive assistance tailored to individual users. In intelligent user interfaces we investigate issues in animated pedagogical agents, user modeling, and affective reasoning. In natural language processing we investigate issues in explanation generation and narrative prose generation. We have begun exploring the use of statistical learning techniques in each of these areas.

Applications
Our work revolves around the design, implementation, and empirical evaluation of cognitively-grounded educational software and natural language systems. In the course of our research over the past decade, we have built several intelligent tutoring systems and empirically studied them, some in the lab and some in the classroom. We have also created several natural language systems, including a narrative prose generator and a legal document generator. Because of insights gained from the practical considerations of fielding software for real users, we are interested in stress-testing our technologies in education, training, and large-scale natural language applications.

Funding
Support for our work has been provided by the National Science Foundation through the NSF Human-Computer Interaction Program, the NSF Learning & Intelligent Systems Program, the NSF Research on Education Policy & Practice Program, the NSF Advanced Learning Technologies Program, and the NSF CreativeIT Program. Additional support has been provided by Novell, IBM, the North Carolina Biotechnology Program, and the William S. Kenan Institute for Engineering, Technology, and Science. We are also grateful to Apple, IBM, and Microsoft for equipment and software donations.