An automatic approach for analysing online discussion forums via text mining

Tak-Lam Wong, Wai Shing Ho, Jeff Tang and Fu Lee Wang
Caritas Institute of Higher Education
Hong Kong SAR, China

Gary Cheng
The Hong Kong Institute of Education
Hong Kong SAR, China

Online discussion forums have been widely used in distance learning and blended learning for developing critical thinking and communication skills. To handle the increasing number of posts in discussion forums, we developed an automatic approach for analysing online discussion data based on a text mining technique. One characteristic of our approach is that text clustering was applied to automatically extract the arguments from posts on a discussion topic. Similar arguments from different users can be grouped together for better analysis by teachers or students. We have conducted a case study using a discussion forum on a course for in-service teachers to evaluate the effectiveness and usefulness of our approach.