Due to technological advances and the Internet, more and more educational data are becoming available. The data are stored in different formats, with varying levels of structure (structured, semi-structured, unstructured), and in different types of databases (relational, graph, etc). In this project we are researching: "How can a learning analytics architecture be built and applied to address challenges in higher education?" The architecture will have components for data collection and storage, analysis and reporting. The research emphasis is on interoperability, for easier combination of different educational datasets from various sources and in different formats. The merge of datasets can lead to more useful data analysis, and thus more useful insights. Semantic technologies is used to address interoperability aspects, with emphasis on technical and semantic aspects. Legal and organizational interoperability is also of essence.
As a proof of concept, the architecture will be evaluated by addressing real-world challenges, informed by stakeholder needs. In this research we are trying to identify, for example, how students succeed in higher education. This research is conducted at the University of Bergen, analyzing data from sources such as Learning Management Systems and Student Information Systems. The goal is to give support to students and instructors to improve learning for the individual student by, for example, providing visualization ("dashboards") that can show information about learning in a course, and to develop systems that offer individualised recommendations (e.g., in terms of study techniques, study resources, or courses).