ARGUNAUT - An Intelligent Guide to Support Productive Online Dialogue
ARGUNAUT (2005-2008) investigated awareness and feedback mechanisms for facilitating interaction in e-discussion environments, and the use of graphical argumentation maps and artificial intelligence components.
The aim of ARGUNAUT was to develop tools that assist tutors in visualising e-discussions as an aid for their moderating tasks ("awareness tools"). The support of an advanced artificial intelligence based system will enable tutors to concentrate on tasks that require their unique human capabilities.
The specific objectives of the project were:
• To study how tutors use visualisations to intervene in graphically represented (mapped) e-discussions and other ICT-supported collaborative learning activities
• To define criteria according to which these discussions can be assessed and improved
• To develop a computational environment which, based on recordings of past discussions and additional information, supports the work of online course moderators and helps them to improve their practice.
The tools developed in ARGUNAUT have the potential to significantly increase the educational productivity of e-learning discussions. They have been tested in four educational settings: higher, secondary, vocational and professional education.
The novel tools have been tailor-made for working on two existing discussion-supporting platforms – the results of past EU-sponsored projects: DUNES/Digalo and Cool Modes. However, ARGUNAUT components are easily adaptable to a wider range of platforms and environments.
The target group is education professionals at all levels, working with argumentative practices in teaching and learning.
Country: France, Germany, Greece, Israel, United Kingdom
Coordinator: Hebrew University, Israel, http://ncjw-res.mscc.huji.ac.il/default_e.asp
Target groups: researchers, teachers
Topic: Information technology, Education, Other
Start year: 2005
End year: 2008
Contact person: Baruch Schwarz, Hebrew University, Israel, info (at) argunaut.org
The ARGUNAUT consortium consisted of two main bodies – a technological team and a pedagogical team.
The pedagogical team analysed discussion maps generated in classrooms in different countries during the DUNES project and continued to enrich the discussion database with new discussions during the project period.
The technological team developed the artificial intelligence components and other tools needed.
Prior to ARGUNAUT, the characteristics and demands of the role of the moderator/tutor in e-learning had not been fully studied, especially in synchronous discussions using environments displaying graphical argumentation maps. Moreover, appropriate tools to assist moderators in these duties had not been developed to any satisfactory level.
ARGUNAUT aimed at unifying awareness and feedback mechanisms for working in e-discussion environments. At the heart of the AI components proposed, an off-line analysis mechanism based on machine learning techniques – the "deep loop" – takes selected human-annotated examples and generalises them to indicators that can in turn be used for awareness feedback. The evaluation scenarios rely on the analysis of discourse arising from this setting.
The DUNES project developed an e-learning tool that enables teachers to visualise students' arguments in discussion maps. ARGUNAUT took another step and included artificial intelligence components in the tool.
The vision is to provide a discussion platform with the ability to assist the moderator in the moderating process by analysing the discussion map on-line, by highlighting trends, dynamics, and potential threats to the discussion direction.
The software is now available to download from the ARGUNAUT website: http://www.argunaut.org/publications/downloads