Detection of Confusion in Online Learners Using Electroencephalography (EEG)

The purpose of this study is to conduct a systematic study on the use of a number of popular machine learning algorithms for classifying confusion in students based on their brainwaves as acquired via electroencephalographyElectroencephalography (EEG) when participating in online learning. The highest accuracy for intra-subject classificationIntra-subject classification was obtained by RF at 93.75% with an average of 79.48% while for inter-subject classificationInter-subject classification, the highest accuracy was also obtained using RF at 65.78% with an average of 63.65%. Our results also show that better classification accuracies were obtained when the confusion detected based on student-defined labels instead of observer-defined labels.