Software Agents and the Future Communications Network

Wayne Wobcke
Department of Information Systems
University of Melbourne

Friday, 30th March 2001
0930 - 1100

RMIT University Function Room (near Kaleide Theatre)
Building 8, level 2, RMIT
Swanston Street, Melbourne.

Abstract:

The future communications network aims to seamlessly combine the current fixed-line telephone, mobile telephone and internet in one framework, opening up many possibilities for new integrated services and new applications for software agents.  This talk will describe two aspects of work of the speaker and his colleagues at British Telecom: (i) the Zeus platform for developing collaborative multi-agent systems of the type that will be required for large-scale applications in business-business e-commerce, and (ii) the Intelligent Personal Assistant, an integrated system of software agents each of which aims to help its user in one area of time, information or communication management.  General research issues in user interface design and in machine learning related to the development of realistic and usable personal assistants will be discussed.

Biosketch:

Wayne Wobcke is a Senior Research Fellow in the Department of Information Systems at the University of Melbourne.  After graduating with a PhD in Computer Science (1989) from the University of Essex, he taught for 9 years as a lecturer in the Basser Department of Computer Science at the University of Sydney, before spending two years working in the Intelligent Systems Research Group at British Telecom Laboratories in the U.K.  His research interests centre on the application of logic in Artificial Intelligence to formalize problems in knowledge representation, natural language processing and agent-based computing.  Whilst at BT, he contributed to the development of three prototype systems: (i) a personal diary assistant combining fuzzy scheduling and multi-agent negotiation, (ii) an information agent providing natural language access to Yellow Pages, and (iii) a scheduling agent for coordinating a system of interface agents.  His current work concerns the investigation of specification and verification techniques for agent-based systems, and techniques for plan and intention recognition with rational agent architectures.