A REVIEW OF CLASSIFICATION TECHNIQUES FOR THE PREDICTION OF HUMAN EMOTION THROUGH HEART RATE AND EYE MOVEMENT

Abstract

Classification is the technique applied in data mining to form groups under specified class labels. Classification is a supervised type of machine learning. In this paper, five familiar classification techniques, Fuzzy Inference System (FIS), Adaptive-Network-based Fuzzy Inference System (ANFIS), Convolutional Neural Network (CNN), Support Vector Machine (SVM), and ANFIS-PLS are compared for accuracy of classification of images. Therefore, we present a comparative study of different classification techniques applied in predicting human emotion through heart rate and eye movement. This paper aims to describe and review the differences in classification techniques and the suitability of each classification to apply in the prediction of human emotion through heart rate and eye movement. Based on the literature survey from four databases: ACM, IEEE, Scopus, and Science Direct; some articles have been reviewed. The studies found that the Adaptive neuro-fuzzy inference system (ANFIS) is the most adopted model due to the motivation mechanism applied. A fundamental review of the selected technique is presented for introduction purposes. A brief comparison with other classifiers, the main advantages and drawbacks of these classifiers are discussed as well.