A Learning Analytics Approach to Identify Students at Risk of Dropout: A Case Study with a Technical Distance Education Course

Abstract

Contemporary education is a vast field that is concerned with the performance of education systems. In a formal e-learning context, student dropout is considered one of the main problems and has received much attention from the learning analytics research community, which has reported several approaches to the development of models for the early prediction of at-risk students. However, maximizing the results obtained by predictions is a considerable challenge. In this work, we developed a solution using only students’ interactions with the virtual learning environment and its derivative features for early predict at-risk students in a Brazilian distance technical high school course that is 103 weeks in duration. To maximize results, we developed an elitist genetic algorithm based on Darwin’s theory of natural selection for hyperparameter tuning. With the application of the proposed technique, we predicted the student at risk with an Area Under the Receiver Operating Characteristic Curve (AUROC) above 0.75 in the initial weeks of a course. The results demonstrate the viability of applying interaction count and derivative features to generate prediction models in contexts where access to demographic data is restricted. The application of a genetic algorithm to the tuning of hyperparameters classifiers can increase their performance in comparison with other techniques.

Towards Predicting Student’s Dropout in University Courses Using Different Machine Learning Techniques

“…Feature extraction refers to the compilation of a subset of unique predictive features for the predictive problem in modelling, also known as attribute selection. The process helps to identify relevant attributes in the dataset that contribute to the accuracy of the prediction model, for instance, by the last activity on the student's corresponding course to predict a drop out [4]. Strategies effective for one kind of datasets may not be sufficient for another one.…”

“…The dropout has a negative impact on educational organizations as well as stakeholders. Even though it has been the focus of many research, e-learning dropout rates generally tend to be higher than face-to-face education [4].…”

Towards an Approach Based on Adjusted Genetic Algorithms to Improve the Quantity of Existing Data in the Context of Social Learning

“…; Kappel, K.; Aguiar, M.; Araújo, R.M. ; Munoz, R.; Villarro-el, R.; Cechinel, C (2020) propose to exploit genetic algorithms to predict at-risk students in a Brazilian university [12]. The solution developed includes the integration of the different interactions of learners within the virtual learning environment.…”

Curricular Analytics to Characterize Educational Trajectories in High-Failure Rate Courses That Lead to Late Dropout

“…In recent years, both researchers and practitioners have been applying learning analytics to support curriculum understanding and improvement [1,2]. Worldwide participation in higher education has grown around the world [3].…”

“…Different methods have been used to understand and predict the dropout phenomenon. Qualitative techniques include surveys and in-depth interviews [7,8], whereas quantitative techniques include a broad range of statistical and data mining methods, such as neural networks [9], multinomial logistic regression [10,11], genetic algorithms [2], and event history analysis [6,12]. The application of longitudinal analysis techniques [12] has also shown that the risk of dropout changes over time and that dropout does not generally occur as a spontaneous event, but rather as a consequence of a particular process [13].…”