Track: Undergraduate Student Paper Competition
Abstract
This work presents a real-time biofeedback tool that uses wearables and Internet of Things (IoT) for applications in education. Using wearables (electroencephalography helmet, smart bands) and a Raspberry Pi, signals were integrated in real-time. Moreover, a three-class random forest (RF) classification machine learning (ML) algorithm, based on the aforementioned signals, which predicted a student's mental fatigue (none, moderate and extreme fatigue) with 92.69% average correct classification percentage using 5-fold cross-validation (CV). The system can evaluate a student's performance under different learning modalities, in addition, it can show different content types to students, depending on the professor's necessities. In the current work, vehicle signals were integrated for teaching automotive engineering.