Title: Comparing ANN, LDA, QDA, KNN and SVM algorithms in classifying relaxed and stressful mental state from two-channel prefrontal EEG data
Authors: Subhrangsu Aditya; D.N. Tibarewala
Addresses: School of Bio Science and Engineering, Jadavpur University, Kolkata, 49, Naskarpara Road, Santoshpur, Kolkata 700075, India ' School of Bio Science and Engineering, Jadavpur University, 61B, Sardar Shankar Road, Kolkata 700029, India
Abstract: This paper attempts to explore the feasibility of classifying relaxed and stressful mental states based on two-channel prefrontal EEG signal from 35 healthy human subjects. Specific objective of this paper is to explore the best choice of features and compare the performance of various feature classification algorithms suitable for this purpose. Here, we included different bivariate features in time domain and frequency domain and compared the classification performance of artificial neural network, linear discriminant analysis, quadratic discriminant analysis (QDA), K nearest neighbour and support vector machine algorithms. Common spatial patterns (CSP) algorithm was used successfully for feature reduction. Best classification performance (99.69%) was observed with the QDA classifier taking cross-correlation estimate as feature. We also explored the effect of combining different kinds of features, effect of varying the number of features on classifier performance, robustness of the chosen methods against in inter-individual variability and the feasibility of developing subject-independent classifiers.
Keywords: stressful mental states; mental stress; prefrontal EEG signals; electroencephalograms; cross-correlation; cross-covariance; cross-spectral density; cross-coherence; artificial neural networks; ANNs; quadratic discriminant analysis; QDA; K nearest neighbour; KNN; support vector machines; SVM; common spatial patterns; CSP; feature selection; feature classification.
International Journal of Artificial Intelligence and Soft Computing, 2012 Vol.3 No.2, pp.143 - 164
Received: 12 Jan 2012
Accepted: 25 Feb 2012
Published online: 29 Nov 2014 *