EEG-based fuzzy cognitive load classification during logical analysis of program segments

The paper aims at designing a novel scheme for cognitive load classification of subjects engaged in program analysis. The logic of propositions has been employed here to construct program segments to be used for cognitive load analysis and classification. Electroencephalogram signals acquired from the subjects during program analysis are first fuzzified and the resultant fuzzy membership functions are then submitted to the input of a fuzzy rule-based classifier to determine the class of the cognitive load of the subjects. Experimental results envisage that the proposed classifier has a good classification accuracy of 86.2%. Performance analysis of the fuzzy classifier further reveals that it outperforms two most widely used classifiers: Support Vector Machine and Naive Bayes classifier.