1 A Human-Computer Dialogue System for Educational Debate: A Computational Dialectics Approach Tangming Yuan. Computer Science Department, University of Akureyri, Sólborg, 600 Akureyri, Iceland yuan@unak.is David Moore and Alec Grierson. Innovation North, Leeds Metropolitan University, Beckett Park, Leeds LS6 3QS, United Kingdom d.moore@leedsmet.ac.uk, a.j.grierson@leedsmet.ac.uk Abstract: This paper reports research concerning issues involved in adopting a computational dialectics approach to develop a human-computer dialogue system for educational debate. In particular, we propose a dialogue model and a set of computationally usable strategic heuristics to enable the computer to engage its users in debate on a controversial issue. The system operationalising the proposed dialogue model and strategy has been fully implemented. It has also been subject to initial evaluation exercises. The results of the evaluation are essentially favourable. It is anticipated that the work reported in this paper will contribute toward the development of human-computer dialogue in general and of computer-based educational debate in particular, and help to illuminate research issues in the field of dialectics itself. Keywords: computational dialectics; educational human-computer debate; dialogue model; computational strategic heuristics INTRODUCTION The recent development of computer based learning (CBL) systems and the emergence of the World Wide Web and the Internet has changed the study life of many people (e.g. Beck, & Stern, 1999; Cumming, & McDougall, 2000; Cunningham-Atkins, Powell, Moore, Hobbs, & Sharpe, 2004; Grasso, & Leng, 2003). However, an assumption underlying many of these computer based educational systems is that the computer does all the informing, the student being merely a passive receiver of the information. The teaching interaction, that is, may become unduly didactic (Moore, 2000; Baker, 2000). There is therefore a need for more dialogue within computer based learning systems. There are many types of dialogue interactions in which people reason together, such as debate, persuasion, inquiry and information-seeking (Walton, & Krabbe, 1995). The debating style of dialogue interaction is argued by Maudet and Moore (2001) to be important in critical thinking and developing debating and reasoning skills, and is also suggested by Pilkington and Mallen’s (1996) educational discourse analysis to be effective and to have rich educational benefit. A particular concern with our research therefore is to investigate issues surrounding a computer based system for educational debate. The educational benefit of such a debating system seems clear, in that the importance of discussion and debate in education is frequently stressed at primary (e.g. National Curriculum Council,