A Linked-Data-Driven Web Portal for Learning Analytics: Data Enrichment, Interactive Visualization, and Knowledge Discovery

This paper presents a Linked-Data-driven Web portal for the field of learning analytics. The portal allows users to browse the linked datasets and explore data about researchers, conferences, and publications. Additionally, users can interact with various dynamic visualization applications and perform analysis, e.g., study temporal change of research trends. Based on the provided datasets on Learning Analytics and Knowledge (LAK) and Educational Data Mining (EDM), we enriched the data with geospatial locations of research institutes, topics extracted from papers, and the expertise of researchers. The interactive modules of the Web portal are then designed and implemented using the enriched RDF data. The implemented modules can be divided into two groups. The first group is concerned with providing dynamic and interactive visualization of the data, such as the modules of Conference Participants and Reference Map. The modules in the second group are designed for more advanced analysis and discovery of new knowledge, such as the modules of Scholar Similarity and Reviewer Recommendation. The modules have been designed following a loosely coupled, modular infrastructure, and can be easily migrated and reused in other projects.