Visual Attention Score and Fatigue Level Measure of Students through Eye Analysis–Machine Learning Approach

The new education paradigm changes the classrooms to E-Learning platforms. The student engagement monitoring is a key problem in E-Learning. This work put forward an efficient mechanism to measure the visual attention of a learner by analyzing various face attributes and the fatigue level. Eye closure is considered as an indicator of fatigue in humans. The novel architecture employs machine learning algorithms to predict the fatigue through bi-state eye classification. Different experimentation are done using various classifiers and feature extraction techniques. The accuracy of the best model derived reaches up to 93.6% through principal component analysis based feature extraction and random forest classifier. The testing of the model is performed on different datasets and on real-time video. The results indicate the efficacy of the model in analyzing the visual attention score during learning.