Text Mining Q&A Websites for Supporting Course Design and Curriculum Development in Higher Education

Authors: Arash, Joorabchi; Michael, English; Abdulhussain E., Mahdi
Year: 2015
Venue: In Proceedings of the 9th International Technology, Education and Development Conference (INTED 2015), pp. 1170-1178 , Madrid, Spain, 2-4 March 2015.
Link: http://library.iated.org/view/JOORABCHI2015TEX
Product of the Action: No

In this work we propose a novel text mining approach for assisting academic staff in their course design and curriculum development activities. The proposed approach is based on mining the textual content of technical and academic Q&A websites and online forums to find the most up-to-date and challenging topics whose inclusion in the curriculum would enhance the quality of related courses offered in higher education institutes. In specific, we have applied this method to a well-established online forum in the field of computer science, StackOverflow.com, which covers a wide range of computer programming topics, and currently contains over fifteen million questions and answers. We have automatically analysed the content of these Q&A’s to determine their topics and broad categories according to the Wikipedia serving as a crowd-sourced classification scheme. We present the results of our computer-assisted analysis in terms of the most frequently occurring subjects and topics and discuss the insight gained from the findings of our analysis.