Global learning network analytics to enhance PLN understanding

In the 21^st century learning increasingly happens on the social web. Learning has evolved into an interactive social process producing large amounts of data across a multitude of inhomogeneous systems. To identify the role of individual actors and groups of actors the whole global learning network needs to be analyzed. The work presented in this paper ingests learning data into a cloud hosted distributed temporal graph model with a supporting distributed processing framework to calculate global graph metrics. The presented simple architectural approach builds upon the xAPI specification to ensure compatibility and flexibility. Based on the global graph metrics we can detect communities and identify information brokers. This information enhances the understanding of the learner's personal learning network and its development over time. It contributes to ongoing efforts to guide learners' through the tangled undergrowth of the global social learning network towards individuals and communities relevant to their interests, skills and aptitudes.