abstract
- Today, the growing technological turnaround from ''big data'' to ''big analytics'' has made it essential for process scientists and analysts to leverage more powerful techniques in analyzing and understanding the different types of data and formats, otherwise allied to the notion of''semantic data engineering Technological advancement has provided data mining solutions capable of processing the data/information in formats that are conceptually comprehended by humans and computers in real-Time or realworld settings, and facilitating building of systems and applications that inclusively manage the information or data they contain (machine-understandable systems). For this purpose, this study introduced a ''semantic-based process mining'' technique capable of discovering abstract or useful information or models from event logs of a learning process (educational domain), and then subsequently used to predict users' patterns through the semantically-inclined modeling and exploration of the discovered process models. Technically, the study focused on improving the results/outcomes of the learning process and mining using the set of proposed algorithms and model that incorporates semantic annotation (labeling), semantic representations (ontology), and semantic reasoning (inference modules or reasoner) of the discovered models. In turn, the outcome of the method shows that data processing and models analyses provided by traditional process mining techniques in education can be enhanced by adding ''semantic information'' (properties description) to the discovered models or events about the processes and domain.