Design of a computational model for social learning support and analystics
Neil Yuwen Yen, Jason Chi-Shun Hung, Chia‐Chen Chen, Qun Jin · Computers in Human Behavior · 2018
Conventional online learning typically allows an instructor to deliver instruction to students via a predefined curriculum and within a fixed knowledge structure (i.e., explaining the instructional subject). With the dramatic growth of social media technology and correlated data aggregation, some sort of instant knowledge is obtained by daily users. An emerging type of knowledge (i.e., social knowledge) has been identified and may lead to self-paced learning from social networks, which is simply defined as social learning. This article points out three important issues for social learning, namely, knowledge retrieval via temporal social factors, and the connection between social network and the knowledge domain. Two significant automation mechanisms, lecture generation for self-regulated learning and influencing domain computation for opportunity finding, are suggested to facilitate the process of social learning. A prototype system based on Elgg was implemented, sourced by a federated repository that has stored and shared more than 1.5 millions transactions (e.g., content, interactions, etc.). We conclude that timely social knowledge (or crowdsourcing results) can be widely applied in the next era of online learning environment. Findings through the statistical analysis are prospective to support understanding of phenomenon of social learning and design of future learning platform for followup researchers.