Kovanović, V., Joksimović, S., Gašević, D., & Hatala, M.
Proceedings of the Learning Analytics and Machine Learning (LAML) Workshop, March 25, 2014, Indianapolis, IN USA.
Publication year: 2014


In this paper we present the results of an exploratory study that examined the use of text mining and text classification for the automation of the content analysis of discussion transcripts in the context of distance education. We used Community of Inquiry (CoI) framework and focused on the content analysis of the cognitive presence construct given its central position within the CoI model. Our results demonstrate the potentials of proposed approach; The developed classifier achieved 58.4% accuracy and Cohen’s Kappa of 0.41 for the 5-category classification task. In this paper we analyze different classification features and describe the main problems and lessons learned from the development of such a system. Furthermore, we analyzed the use of several novel classification features that are based on the specifics of cognitive presence construct and our results indicate that some of them significantly improve classification accuracy.