Abstract

This article introduces LANSE, an innovative Learning Analytics tool tailored for Learning Management Systems, with the primary goal of identifying student behaviors to predict risks of dropout and failure. The tool uses a cloud-based architecture that supports comprehensive data collection, processing, and visualization. In order to detect students at-risk, the tool offers automated models trained by machine learning algorithms that provide weekly predictions about the risk of the students, together with visualizations about their interactions inside the course. The performances of the models for predicting students at-risk of dropout and failure align with the state-of-the-art in the existing literature. Presently implemented in distance learning courses, initial feedback suggests that the tool effectively optimizes workloads and students behavior tracking. Challenges encountered include ensuring privacy compliance, effective data management, and maintaining real-time processing and security measures. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.