An intelligent agent in web-based argumentation

The emergence of the Internet and World Wide Web (WWW) has triggered the development of computer supported collaborative; argumentation (CCA) tools. However, because of the difficulty of dealing with semantic issues, understanding arguments is still a major difficulty in designing and implementing these tools. In this paper, we propose a new approach to alleviate this problem by using the technology of a linguistics corpus in natural language understanding. From our survey and research, we find that computer-based arguments contain many regular patterns that occur with high frequency. Further more, many argumentation strategies and rhetorical methods are related to these regular patterns. Arguments can be partially or locally understood through these regular patterns, and a system can provide students with advice by interpreting these patterns. We implemented this approach on the platform of HyperNews. All technology details; art: hidden from users. Users experience an intelligent agent providing support for argumentation strategies and rhetorical methods during the process of web-based argumentation.