e-LION: Data integration semantic model to enhance predictive analytics in e-Learning

In the last years, Learning Management systems (LMSs) are acquiring great importance in online education, since they offer flexible integration platforms for organising a vast amount of learning resources, as well establishing effective communication channels between teachers and learners, at any direction. These then attracting an increasing number users that continuously access, download/upload resources interact each other during their teaching/learning processes, which is even accelerating by breakout COVID-19. this context, academic institutions generating large volumes learning-related data can be analysed supporting lesson, course or faculty degree planning, administrations university strategic level. However, managing such data, usually coming from multiple heterogeneous sources with attributes sometimes reflecting semantic inconsistencies, constitutes emerging challenge, so require common definition schemes to easily fuse them, aim efficiently feeding machine models. regard, web technologies arise useful framework multi-source e-learning allowing consolidation, linkage advanced querying systematic way. With motivation, e-LION (e-Learning Integration ONtology) model proposed first time work operate consolidation approach different knowledge-bases, hence leading enrich on-top analysis. For demonstration purposes, ontological populated real-world private public LMSs referring courses Software Engineering University Malaga (Spain) Open Learning. set four case studies worked validation, comprise advance predictive modelling time-series forecasting students’ interactions according final grades, generation SWRL reasoning rules student’s behaviour classification. The results promising lead possible use mediator scheme new future models domain e-learning.