Emotions play a crucial role in human communication and the Brain-Computer Interfaces (BCI) aid in emotion-based communication, similar to the way a mobile device aids in text-based communication. Emotions are expressed in myriad ways, including verbal, non-verbal and physiological signals. Most BCI systems accomplish this by using electroencephalography (EEG) signals. Before BCI systems are employed to practical use, efficient algorithms need to be developed in order to maximize efficiency. This paper proposes a method to map several emotional states using EEG signals collected from the publicly available Dataset for Emotion Analysis using Physiological signals (DEAP). It presents a comparative analysis of two different classification algorithms along with two different dimensionality reduction techniques. Linear Discriminant Analysis (LDA) and Principal Component Analysis (PCA) are used in tandem with Support Vector Machine (SVM) and K-Nearest Neighbours (KNN). The objective of this paper is to weigh the results to find the most promising classifier and dimensionality reduction technique. The average accuracy of binary classification using KNN-LDA for each valence, arousal, dominance and liking was 97.98%, 96.21%, 98.24% and 96.19% respectively.

Emotion classification, EEG, Physiological signals, Signal processing, Pattern classification, Affective computing, DEAP Dataset, Machine Learning, Brain-Computer Interfaces (BCI), Electroencephalogram (EEG), Linear Discriminant Analysis (LDA), K-Nearest Neighbours (KNN), Support Vector Machine (SVM), Principal Component Analysis (PCA).

[1] T. D. Kemper and R. S. Lazarus, “Emotion and Adaptation,” Contemporary Sociology, vol. 21, no. 4. p. 522, 1992.
[2] J. Posner, J. A. Russell, and B. S. Peterson, “The circumplex model of affect: An integrative approach to affective neuroscience, cognitive development, and psychopathology,” Development and Psychopathology, vol. 17, no. 03. 2005.
[3] Y. Liu, O. Sourina, and M. K. Nguyen, “Real-Time EEG-Based Human Emotion Recognition and Visualization,” 2010 International Conference on Cyberworlds. 2010.
[4] R. Plutchik, “The Nature of Emotions: Clinical Implications,” Emotions and Psychopathology. pp. 1–20, 1988.
[5] D. Garrett, D. A. Peterson, C. W. Anderson, and M. H. Thaut, “Comparison of linear, nonlinear, and feature selection methods for EEG signal classification,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 11, no. 2, pp. 141–144, Jun. 2003.
[6] M. Soleymani, M. Pantic, and T. Pun, “Multimodal emotion recognition in response to videos (Extended abstract),” 2015 International Conference on Affective Computing and Intelligent Interaction (ACII). 2015.
[7] R. Nivedha, M. Brinda, D. Vasanth, M. Anvitha, and K. V. Suma, “EEG based emotion recognition using SVM and PSO,” 2017 International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT). 2017.
[8] A.S. Mali, A.A. Kenjale, P.M. Ghatage, A.G. Deshpande, "Mood based Music System", International Journal of Scientific Research in Computer Science and Engineering, Vol.6, Issue.3, pp.27-30, 2018
[9] Afzal Ahmad, Mohammad Asif, Shaikh Rohan Ali, "Review Paper on Predicting Mood Disorder Risk Using Machine Learning", International Journal of Scientific Research in Computer Science and Engineering, Vol.7, Issue.1, pp.16-22, 2019
[10] S. Koelstra et al., “DEAP: A Database for Emotion Analysis; Using Physiological Signals,” IEEE Transactions on Affective Computing, vol. 3, no. 1. pp. 18–31, 2012.
[11] Russell, J. A. (1980), “A circumplex model of affect,” Journal of Personality and Social Psychology, 39, December, 1161-1178.