Fleshing Out Learning Analytics and Educational Data Mining with Data and ML Pipelines

Advances in data collection techniques and data storage devices have enabled the collection and persistent storage of data flowing into different operational systems from various sources. Educational systems is not an exception to this rule, and as such continuously generated streams of data related to the interaction among students. Instructors and learning materials in the context of the learning processes, are constantly accumulating inside different data repositories. Exploratory Data Analysis, Data Mining and Machine Learning as well as Big Data Analytics techniques can be used to make sense out of this multitude of data in order to offer benefit to educational institutes from a range of levels. In this paper, we present an automated educational data mining and learning analytics approach that relies on Machine Learning (ML) and Data pipelines which has as their goal to improve both the learning experience of students and the teaching experience of tutors, as well as the institutional strategic view of the educational structures. By exploring and analyzing all the collected data, we build models that explain and assess the effectiveness of the learning environment, so as to address the needs of the aforementioned members of these institutions. In this underlying framework, we also present and evaluate a case study that focuses on analyzing educational data for assessment purposes from a big data infrastructure of a distance learning and online Higher Education Institute.