Adopting Trust in Learning Analytics Infrastructure: A Structured Literature Review

George-Petru Ciordas-Hertel (DIPF - Leibniz Institute for Research and Information in Education, Germany)

Jan Schneider (DIPF - Leibniz Institute for Research and Information in Education, Germany)

Stefaan Ternier (Open Universiteit, The Netherlands)

Hendrik Drachsler (DIPF - Leibniz Institute for Research and Information in Education Frankfurt am Main, Germany, Johann Wolfgang Goethe-Universitat, Germany and Open Universiteit, The Netherlands)

Abstract: One key factor for the successful outcome of a Learning Analytics (LA) infrastructure is the ability to decide which software architecture concept is necessary. Big Data can be used to face the challenges LA holds. Additional challenges on privacy rights are introduced to the Europeans by the General Data Protection Regulation (GDPR). Beyond that, the challenge of how to gain the trust of the users remains. We found diverse architectural concepts in the domain of LA. Selecting an appropriate solution is not straightforward. Therefore, we conducted a structured literature review to assess the state-of-the-art and provide an overview of Big Data architectures used in LA. Based on the examination of the results, we identify common architectural components and technologies and present them in the form of a mind map. Linking the findings, we are proposing an initial approach towards a Trusted and Interoperable Learning Analytics Infrastructure (TIILA).

Keywords: GDPR, big data, data protection, education, infrastructure, learning analytics, privacy, software architecture, trust

Categories: D.2.11, D.2.12, K.3, L.3.6