Title: Toward enhancing collaborative learning groups formation in Q&A website using tag-based next questions predictor

Authors: Mohammad Sadegh Rezaei; Hossein Bobarshad; Kambiz Badie

Addresses: School of Engineering, Department of Network Sciences and Technology, Faculty of New Sciences and Technologies, University of Tehran, Tehran, Iran ' School of Engineering, Department of Network Sciences and Technology, Faculty of New Sciences and Technologies, University of Tehran, Tehran, Iran ' Iran Telecom Research Center, Tehran, Iran

Abstract: The advent of informal social learning environment has provided the necessary platform for realising of lifelong learning. The basic platform of learning in an informal environment is the learning groups called online Community of Practice (oCoP). This study proposes a predictor for predicting the topics most suitable for the formation of oCoPs for the users adopting the roles of co-learner and mentor in a shared domain perspective. Proposed predictor operates based on Naive Bayes prediction and collaborative filtering and uses educational social tags for realising the learners' roles and behaviour. The predictor is evaluated with the dataset of the website StackOverflow. The results show that the proposed predictor can predict the next question of learners with high accuracy. Therefore, the system can facilitate the formation of suitable learning groups around the predicted users' interest topics.

Keywords: social learning network; Q&A website; online community of practice; informal learning; learning topic prediction; collaborative filtering; collaborative learning; educational social tag.

DOI: 10.1504/IJTEL.2019.102570

International Journal of Technology Enhanced Learning, 2019 Vol.11 No.4, pp.441 - 462

Received: 22 May 2018
Accepted: 04 Sep 2018

Published online: 12 Apr 2019 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article