Next learning topic prediction for learner's guidance in informal learning environment

Abstract

Estimating the learning needs of learners in a Social Learning Network (SLN) is very important in proper planning for improving learning space. This paper presented a predictor to estimate the learning needs of learners in SLNs. In Question & Answer Networks, estimating the need for learning means estimating the future subject of the question. The significance of the similarity of the sequence of previous learning subjects with the future subjects of learners is one of the most important areas for estimating the subject of future learning. Hence, this predictor estimates the next learning subject based on the similarity of the subjects about which the learner asks questions. The estimation method introduced in this study is based on the Bayesian solution method. The performance of this method was evaluated in the dataset extracted from one of the most widely used SLNs. The results showed that the proposed method was able to detect future tag of each learner with 78% precision in the informal learning environment using the tags of the questions asked by learners.