An ontology‐based framework of assessment analytics for massive learning

Abstract

AbstractAnalyzing learning traces is presently highly required in e‐learning environment. Several communities have been developed to address this need, such as those of Learning Analytics and Educational Data Mining. The main step of performing a learning analytics process is the educational data collection.Actually, learning environments such as Massive Open Online Course (MOOC) generate a big amount of educational data. They can be divided into assessment data, collaboration data, communication data, and so on. When we focus on assessment, we can launch a new source of data that can be analyzed and hence contribute to the improvement of learning analytics field.In this paper, we explore, investigate and compare the set of learning analytics models in the literature. Then, we study them from assessment point of view. The only current learning analytics model which can support tracking and modeling assessment data is the xAPI data model. For this reason, we study and investigate the xAPI specification from assessment point of view. Based on identified weaknesses of xAPI specification, we propose an enhancement of its data model. This is to support the assessment analytics effectively. We present an ontological model for assessment analytics inspired from the xAPI specification. To validate our approach, we focus on massive learning traces extracted from a real MOOC. Thus, we define and execute the set of proposed steps of preprocessing stage that extracts assessment data from whole learning data. Furthermore, we develop a java semantic web application to convert assessment data extracted to OWL file according to our proposed ontological model for assessment analytics.