TSK-Based Type-2 Fuzzy Analysis of Infrared Spectroscopic Data for Classification of Touch-Induced Affection

Abstract

The paper introduces a novel approach to categorize the hemodynamic response of subjects due to arousal of touch induced affection classes such as Respect, Love, Fondness and Devotion using a TSK-based Type-2 Fuzzy classifier. The main contribution of the paper is to design the novel TSK-based Interval Type-2 Fuzzy classifier to classify the finer changes in affective emotions using the hemodynamic response of a subject, when she comes in contact with her mother, spouse, child and also conveys her prayer to a model/sculpture of God by holding it with her palms. Experiments undertaken reveal that the brain activation patterns varies in different sub-regions over distinct time-windows for individual emotions. Relative performance analysis and statistical validation confirm the superiority of the proposed TSK-based Interval Type-2 Fuzzy classifier. Moreover, the proposed scheme has successfully been applied for assessing subjective sensitivity of healthy as well as psychiatric disordered people.