Exploring approaches to educational data mining and learning analytics, to measure the level of acquisition of student's learning outcome
Diego Buenaño-Fernández, Sergio Luján-Mora
Proceedings of the 8th International Conference on Education and New Learning Technologies (Edulearn 2016), p. 1845-1850, Barcelona (Spain), July 4-6 2016. ISBN: 978-84-608-8860-4. https://doi.org/10.21125/edulearn.2016.1368
(EDULEARN'16g) Congreso internacional / International conference
The Educational Data Mining community website defines educational data mining as: “Educational Data Mining is an emerging discipline, concerned with developing methods for exploring the unique types of data that come from educational settings, and using those methods to better understand students, and the settings which they learn in.” The increase of interactive learning environments such as Learning Management Systems (LMS), Personal Learning Environments (PLE), intelligent tutoring systems, generates large amounts of data. Nevertheless, in the majority of cases, the data stored in virtual platforms, either in files or databases, are underused by teachers, students and institutions. These educational datasets provide multiple opportunities to exploit the power of technology in the teaching-learning process. In addition, these data generate a space for the increase of research in topics related to: evaluation of learning theories, learning technology, learner feedback and support, early warning systems, and the development of future learning applications. Although the tools to analyze the log files stored on the LMS allow to measure, collect, evaluate and present all student data, these do not include artificial intelligence algorithms as a support mechanism for decision. Employing artificial intelligence to LMS logs analysis would provide teachers insight into each individual student's activity. Based on that information, professors could adjust course materials according to student interest and knowledge. Several researches have shown the use of learning analytics, machine learning, data mining techniques and classifiers to predict on the issue of learning outcomes in students. Examples of these technologies are serious game based on Blooms taxonomy, classification techniques among which Naïve Bayes, ontology model for classifying learning outcomes based on Blooms taxonomy. A particular use of the Learning Analytics (LA) is feedback and enhance the learning assessment process. The processes of learning assessment traditional have been complex and often subjective; dependent only invites the vision of the teacher. This has not allowed measure the achievement of skills and abilities in students, these methodologies have mainly assessed cognitive skills. The use of Bloom's taxonomy along with tools of data mining and LA it is suggested to generate knowledge classification criteria. According to this taxonomy, there are six mayor outcomes - Knowledge, Comprehension, Application, Analysis, Synthesis, and Evaluation. This paper presents and survey a summary of the most relevant approaches of the use of techniques and tools of educational data mining and learning analytics, using LMS platform’s in higher education.