Evaluation of Emotions from Brain Signals on 3D VAD Space via Artificial Intelligence Techniques
Abstract
:1. Introduction
2. Materials and Methods
2.1. Dataset and Channels
2.2. Overall Method and Classification
2.3. Classes
2.4. Attributes
2.5. SMOTE
2.6. Classifiers
2.7. Metrics
3. Results
3.1. Segmentation Pre-Study
3.2. Cross-Comparisons of 3D VAD Emotions, Segments, and Classes
4. Discussion
- A: Happy, joyful, useful, powerful, influential, friendly, excited, and inspired
- B: Relaxed, leisurely, untroubled, and satisfied
- C: Disdainful and unconcerned
- D: Angry, hostile, cruel, insolent, hate, and disgusted
- E: Impressed, surprised, thankful, and curious
- F: Protected, humble, and reverent
- G: Sad, depressed, bored, lonely, feeble, discouraged, and discontented
- H: Fearful, frustrated, helpless, pain, humiliated, embarrassed, guilty, and confused
- A&G (Happy, joyful, excited, and powerful and sad, depressed, and feeble)
- B&H (Relaxed, leisurely, and satisfied and fear, pain, and guilt)
- D&F (Angry, hate, and disgusted and protected, humble, and reverent)
- C&E (Disdainful and unconcerned and impressed, surprised, and thankful).
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Paul, S.; Mazumder, A.; Ghosh, P.; Tibarewala, D.N.; Vimalarani, G. EEG Based Emotion Recognition System using MFDFA as Feature Extractor. In Proceedings of the International Conference on Robotics, Automation, Control and Embedded Systems–RACE 2015, Hindustan University, Chennai, India, 18–20 February 2015. [Google Scholar]
- Sternberg, R.J.; Sternberg, K. Cognitive Psychology, 6th ed.; Wadsworth Cengage Learning: Belmont, CA, USA, 2012. [Google Scholar]
- Bălan, O.; Moldoveanu, A.; Leordeanu, M. A Machine Learning Approach to Automatic Phobia Therapy with Virtual Reality. In Modern Approaches to Augmentation of Brain Function; Opris, I., Lebedev, M.A., Casanova, M.F., Eds.; Springer: Cham, Switzerland, 2021; pp. 607–636. [Google Scholar] [CrossRef]
- Zhang, J.; Yin, Z.; Chen, P.; Nichele, S. Emotion recognition using multi-modal data and machine learning techniques: A tutorial and review. Inf. Fusion 2020, 59, 103–126. [Google Scholar] [CrossRef]
- Chikara, R.K.; Chang, E.C.; Lu, Y.C.; Lin, D.S.; Lin, C.T.; Ko, L.W. Monetary Reward and Punishment to Response Inhibition Modulate Activation and Synchronization within the Inhibitory Brain Network. Front. Hum. Neurosci. 2018, 12, 27. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Yu, W.; Ding, S.; Yue, Z.; Yang, S. Emotion Recognition from Facial Expressions and Contactless Heart Rate Using Knowledge Graph. In Proceedings of the IEEE International Conference on Knowledge Graph (ICKG), Nanjing, China, 9–11 August 2020. [Google Scholar] [CrossRef]
- Shabbir, J.; Anwer, T. Artificial Intelligence and its Role in Near Future. arXiv 2015, arXiv:1804.01396. [Google Scholar]
- Doma, V.; Pirouz, M. A comparative analysis of machine learning methods for emotion recognition using EEG and peripheral physiological signals. J. Big Data 2020, 7, 18. [Google Scholar] [CrossRef] [Green Version]
- Kumar, G.S.S.; Arun, A.; Sampathila, N.; Vinoth, R. Machine Learning Models for Classification of Human Emotions Using Multivariate Brain Signals. Computers 2022, 11, 152. [Google Scholar] [CrossRef]
- Abdulrahman, A.; Baykara, M.; Alakus, T.B. A Novel Approach for Emotion Recognition Based on EEG Signal Using Deep Learning. Appl. Sci. 2022, 12, 10028. [Google Scholar] [CrossRef]
- Alakuş, T.B.; Türkoğlu, İ. EEG Based Emotion Analysis Systems. In Türkiye Bilişim Vakfı Bilgisayar Bilimleri ve Mühendisliği Dergisi; Türkiye Bilişim Vakfı: Iistanbul, Turkey, 2018; Volume 15. [Google Scholar]
- Verma, G.K.; Tiwary, U.S. Affect representation and recognition in 3D continuous valence–arousal–dominance space. Multimed Tools Appl. 2017, 76, 2159–2183. [Google Scholar] [CrossRef]
- Verma, G.K.; Tiwary, U.S. Multimodal fusion framework: A multiresolution approach for emotion classification and recognition from physiological signals. NeuroImage 2014, 102, 162–172. [Google Scholar] [CrossRef]
- Russell, J.A.; Mehrabian, A. Evidence for A Three-Factor Theory of Emotions. J. Res. Personal. 1977, 11, 273–294. [Google Scholar] [CrossRef]
- Guendil, Z.; Lachiri, Z.; Maaoui, C. Computational framework for emotional VAD prediction using regularized Extreme Learning Machine. Int. J. Multimed Inf. Retr. 2017, 6, 251–261. [Google Scholar] [CrossRef]
- Özel, P.; Akan, A.; Yılmaz, B. Synchrosqueezing transform based feature extraction from EEG signals for emotional state prediction. Biomed. Signal Process. Control. 2019, 52, 152–161. [Google Scholar] [CrossRef]
- Koelstra, S.; Muehl, C.; Soleymani, M.; Lee, J.-S.; Yazdani, A.; Ebrahimi, T.; Pun, T.; Nijholt, A.; Patras, I. DEAP: A Database for Emotion Analysis using Physiological Signals. IEEE Trans. Affect. Comput. 2012, 3, 18–31. [Google Scholar] [CrossRef] [Green Version]
- Rhif, M.; Abbes, A.B.; Farah, I.R.; Martínez, B.; Sang, Y. Wavelet Transform Application for/in Non-Stationary Time-Series Analysis: A Review. Appl. Sci. 2019, 9, 1345. [Google Scholar] [CrossRef] [Green Version]
- Sifuzzaman, M.; Islam, M.R.; Ali, M.Z. Application of Wavelet Transform and its Advantages Compared to Fourier Transform. J. Phys. Sci. 2009, 13, 121–134. [Google Scholar]
- Öner, İ.V.; Yeşilyurt, M.K.; Yılmaz, E.Ç. Wavelet Analiz Tekniği Ve Uygulama Alanları. Ordu Univ. J. Sci. Tech. 2017, 7, 42–56. [Google Scholar]
- Frank, E.; Hall, M.A.; Witten, I.H. The WEKA Workbench. Online Appendix for “Data Mining: Practical Machine Learning Tools and Techniques (Version: 3.9.6)”, 4th ed.; Morgan Kaufmann: Burlington, MA, USA, 2016. [Google Scholar]
- Watkins, J.C. An Introduction to the Science of Statistics: From Theory to Implementation, Preliminary, ed.; 2016. Available online: http://www.freetechbooks.com/an-introduction-to-the-science-of-statistics-from-theory-to-implementation-preliminary-edition-t1277.html (accessed on 19 April 2023).
- Gupta, S.C.; Kapoor, V.K. Fundamentals of Mathematical Statistics (A Modern Approach), 10th ed.; Sultan Chand & Sons, Educational Publishers: New Delhi, India, 2000. [Google Scholar]
- Henriques, T.; Ribeiro, M.; Teixeira, A.; Castro, L.; Antunes, L.; Costa-Santos, C. Nonlinear Methods Most Applied to Heart-Rate Time Series: A Review. Entropy 2020, 22, 309. [Google Scholar] [CrossRef] [Green Version]
- Safi, M.S.; Safi, S.M.M. Early detection of Alzheimer’s disease from EEG signals using Hjorth parameters. Biomed. Signal Process. Control. 2021, 65, 102338. [Google Scholar] [CrossRef]
- Rihaczek, A.W. Signal Energy Distribution in Time and Frequency. IEEE Trans. Inf. Theory 1968, 14, 369–374. [Google Scholar] [CrossRef]
- Rafiuddin, N.; Khan, Y.U.; Farooq, O. Feature Extraction and Classification of EEG for Automatic Seizure Detection. In Proceedings of the International Conference on Multimedia, Signal Processing and Communication Technologies, Aligarh, India, 17–19 December 2011. [Google Scholar] [CrossRef]
- Rahman, A.; Hossain, F.; Hossain, M.; Ahmed, R. Employing PCA and t-statistical approach for feature extraction and classification of emotion from multichannel EEG signal. Egypt. Inform. J. 2020, 21, 23–35. [Google Scholar] [CrossRef]
- Thongpanja, S.; Phinyomark, A.; Limsakul, C.; Phukpattaranont, P. Application of Mean and Median Frequency Methods for Identification of Human Joint Angles Using EMG Signal. In Information Science and Applications, Lecture Notes in Electrical Engineering; Kim, K.J., Ed.; Springer: Berlin/Heidelberg, Germany, 2015; p. 339. [Google Scholar] [CrossRef]
- Minkowski, L.; Mai, K.V.; Gurve, D. Feature Extraction to Identify Depression and Anxiety Based on EEG. In Proceedings of the 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Virtual Conference, 31 October–4 November 2021. [Google Scholar] [CrossRef]
- Mohamad Ismail, M.R.; Lam, C.K.; Cheng, F.Y.; Sundaraj, K.; Rahiman, M.H.F. Analysis and classification of muscle activity during biceps exercise using MMG signals. In Proceedings of the 2018 4th International Conference on Electrical, Electronics and System Engineering (ICEESE), Kuala Lumpur, Malaysia, 8–9 November 2018; pp. 12–16. [Google Scholar]
- Torun, Y.; Pazarci, Ö. Parametric Power Spectral Density Estimation-Based Breakthrough Detection for Orthopedic Bone Drilling with Acoustic Emission Signal Analysis. Acoust. Aust. 2020, 48, 221–231. [Google Scholar] [CrossRef]
- Li, X.; Song, D.; Zhang, P.; Zhang, Y.; Hou, Y.; Hu, B. Exploring EEG Features in Cross-Subject Emotion Recognition. Front. Neurosci. 2018, 12, 162. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Vijayakumar, S.; Flynn, R.; Murray, N. A Comparative Study of Machine Learning Techniques for Emotion Recognition from Peripheral Physiological Signals. In Proceedings of the 2020 31st Irish Signals and Systems Conference (ISSC), Letterkenny, Ireland, 11–12 June 2020. [Google Scholar]
- Wijayanto, I.; Hadiyoso, S.; Aulia, S.; Atmojo, B.S. Detecting Ictal and Interictal Condition of EEG Signal using Higuchi Fractal Dimension and Support Vector Machine. J. Phys. Conf. Ser. 2020, 1577, 012016. [Google Scholar] [CrossRef]
- Sun, Q.; Wang, Q.; Ji, B.; Wu, W.; Huang, W.; Wang, C. The Cardiodynamicsgram Based Early Detection of Myocardial Ischemia Using the Lempel-Ziv Complexity. IEEE Access 2020, 8, 207894–207904. [Google Scholar] [CrossRef]
- Šapina, M.; Karmakar, C.K.; Kramari, K.; Kosmider, M.; Garcin, M.; Brdari, D.; Milas, K.; Yearwood, J. Lempel-Ziv complexity of the pNN x statistics an application to neonatal stress. Chaos Solitons Fractals 2021, 146, 110703. [Google Scholar] [CrossRef]
- Chawla, N.V.; Bowyer, K.W.; Hall, L.O.; Kegelmeyer, W.P. SMOTE: Synthetic Minority Over-sampling Technique. J. Artif. Intell. Res. 2002, 16, 321–357. [Google Scholar] [CrossRef]
- Rahman, M.; Sarkar, A.K.; Hossain, A.; Moni, M.A. EEG-based emotion analysis using non-linear features and ensemble learning approaches. Expert Syst. Appl. 2022, 207, 118025. [Google Scholar] [CrossRef]
- Vijayarani, S.; Muthulakshmi, M. Comparative Analysis of Bayes and Lazy Classification Algorithms. Int. J. Adv. Res. Comput. Commun. Eng. 2013, 2, 3118–3124. [Google Scholar]
- Platt, J.C. Sequential Minimal Optimization: A Fast Algorithm for Training Support Vector Machines; Microsoft Research, Technical Report MSR-TR-98-14; Microsoft: Redmond, WA, USA, 1998. [Google Scholar]
- Breiman, L. Random Forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef] [Green Version]
- Fawagreh, K.; Gaber, M.M.; Elyan, E. Random forests: From early developments to recent advancements. Syst. Sci. Control. Eng. 2014, 2, 602–609. [Google Scholar] [CrossRef] [Green Version]
- Hühn, J.; Hüllermeier, E. FURIA: An Algorithm for Unordered Fuzzy Rule Induction. Data Min. Knowl. Disc. 2009, 19, 293–319. [Google Scholar] [CrossRef] [Green Version]
- Breiman, L. Bagging predictors. Mach. Learn. 1996, 24, 123–140. [Google Scholar] [CrossRef] [Green Version]
- Adnan, N.; Islam, Z. Optimizing the Number of Trees in a Decision Forest to Discover a Subforest with High Ensemble Accuracy using a Genetic Algorithm. Knowl. Based Syst. 2016, 110, 86–97. [Google Scholar] [CrossRef]
- Chicco, D.; Jurman, G. The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genom. 2020, 21, 6. [Google Scholar] [CrossRef] [Green Version]
- Lipton, Z.C.; Elkan, C.; Naryanaswamy, B. Thresholding Classiffers to Maximize F1 Score. arXiv 2014, arXiv:1402.1892. Available online: https://deepai.org/publication/thresholding-classifiers-to-maximize-f1-score (accessed on 14 September 2022).
- Ahmed, M.Z.I.; Sinha, N.; Ghaderpour, E.; Phadikar, S.; Ghosh, R. A Novel Baseline Removal Paradigm for Subject-Independent Features in Emotion Classification Using EEG. Bioengineering 2023, 10, 54. [Google Scholar] [CrossRef] [PubMed]
- Almanza-Conejo, O.; Almanza-Ojeda, D.L.; Contreras-Hernandez, J.L.; Ibarra-Manzano, M.A. Emotion recognition in EEG signals using the continuous wavelet transform and CNNs. Neural Comput. Appl. 2023, 35, 1409–1422. [Google Scholar] [CrossRef]
- Shu, L.; Xie, J.; Yang, M.; Li, Z.; Li, Z.; Liao, D.; Xu, X.; Yang, X. A Review of Emotion Recognition Using Physiological Signals. Sensors 2018, 18, 2074. [Google Scholar] [CrossRef] [Green Version]
- Chikara, R.K.; Ko, L.-W. Neural Activities Classification of Human Inhibitory Control Using Hierarchical Model. Sensors 2019, 19, 3791. [Google Scholar] [CrossRef] [Green Version]
Class Separation Level | High (H) 1 | Low (L) 1 |
---|---|---|
5 | ≥5 | <5 |
4.5–5.5 2 | ≥5.5 | ≤4.5 |
4–6 | ≥6 | ≤4 |
3–7 | ≥7 | ≤3 |
Segment Name | Windowing (Second) | Segment/Piece |
---|---|---|
N | None | 1 |
Seg I | 30 | 2 |
Seg II | 20 | 3 |
Seg III | 15 | 4 |
Seg IV | 10 | 6 |
Seg V | 5 | 12 |
Seg VI | 3 | 20 |
Seg VII | 2 | 30 |
Seg VIII | 1 | 60 |
Sep. Level | Class | CCI (%) | F1 Score (%) | MCC (%) | ||||||
---|---|---|---|---|---|---|---|---|---|---|
Value | Classifier * | Segment | Value | Classifier | Segment | Value | Classifier | Segment | ||
(5) | A&G | 81.71 | RSS | V | 80.5 | RSS | V | 54.9 | RSS | V |
B&H | 81.98 | RSS | V | 81.9 | RSS | V | 63.8 | RSS | V | |
D&F | 85.68 | RF | V | 85.5 | RF | V | 69.2 | RF | V | |
C&E | 86.84 | RSS | V | 86.8 | RSS | V | 72.9 | RSS | V | |
(4–6) | A&G | 85.13 | RSS | V | 84.3 | RSS | V | 62.3 | RSS | V |
B&H | 91.01 | OF | V | 90.9 | OF | V | 80.7 | OF | V | |
D&F | 92.35 | RF | V | 92.3 | RF | V | 82.9 | RF | V | |
C&E | 96.97 | FURIA | IV | 97.0 | FURIA | IV | 93.9 | FURIA | IV | |
(3–7) | A&G | 90.08 | RSS | V | 89.6 | RSS | V | 72.2 | RSS | V |
B&H | 96.03 | RC | V | 95.9 | RC | V | 85.2 | RC | V | |
D&F | 91.32 | RSS | V | 91.4 | RSS | V | 81.3 | RSS | V | |
(4.5–5.5) | C&E | 98.94 | IBk | III | 98.9 | IBk | III | 97.9 | IBk | III |
Sep. Level | Class | CCI (%) | F1 Score (%) | MCC (%) | ||||||
---|---|---|---|---|---|---|---|---|---|---|
Value | Classifier * | Segment | Value | Classifier | Segment | Value | Classifier | Segment | ||
(5) | A&G | 88.62 | RF | V | 88.6 | RF | IV, V | 77.4 | RF | V |
B&H | 82.60 | OF | V | 82.6 | OF | V | 65.2 | OF | V | |
D&F | 89.94 | OF | V | 89.9 | RF, OF | V | 80.2 | OF | V | |
C&E | 90.31 | OF | V | 90.3 | OF | V | 80.7 | OF | V | |
(4–6) | A&G | 91.45 | RF | V | 91.4 | RF | V | 83.1 | RF | V |
B&H | 93.37 | OF | V | 93.4 | OF | V | 90.8 | FURIA | III | |
D&F | 94.76 | IBk, RC, SMO | III, IV, V | 94.8 | IBk, RC, SMO | III, IV, V | 90.0 | IBk | IV | |
C&E | 97.92 | RC | III | 97.9 | RC | III | 95.9 | RC | III | |
(3–7) | A&G | 95.58 | SMO | IV | 95.6 | SMO | IV | 91.4 | SMO | IV |
B&H | 100.00 | RF, OF | III | 100.0 | RF, OF | III | 100.0 | RF, OF | III | |
D&F | 95.05 | SMO | V | 95.0 | SMO | V | 90.3 | SMO | V | |
(4.5–5.5) | C&E | 97.92 | IBk | III | 97.9 | IBk | III | 95.9 | IBk | III |
Study | Class | Accuracy 1 |
---|---|---|
[12] | 3-class classification (Low, medium, and high) | 63.47 (V) |
69.62 (A) | ||
63.57 (D) | ||
[15] | 2 classes | 73.43% (V) |
72.65% (A) | ||
69.3% (D) | ||
[16] | 2D and 3D emotion models | |
4 emotional states (VA model) | Best accuracy of 79.1%. | |
8 emotional states (VAD model) | Highest accuracy of 93% | |
Our Study | Binary classification for cross-comparisons of 8 emotional states from VAD space (A&G; B&H; D&F; C&E) for unbalanced data | Best accuracy of 98.94% CCI (F1: 98.9% and MCC: 97.9%) for C&E comparison (Seg III and 4.5–5.5 class separation) |
Please see Table 3 for other results. | ||
for balanced data with SMOTE | Best accuracy of 100% CCI, F1 and MCC for B&H comparison (Seg III and 3–7 class separation) | |
Please see Table 4 for other results. |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Işık, Ü.; Güven, A.; Batbat, T. Evaluation of Emotions from Brain Signals on 3D VAD Space via Artificial Intelligence Techniques. Diagnostics 2023, 13, 2141. https://doi.org/10.3390/diagnostics13132141
Işık Ü, Güven A, Batbat T. Evaluation of Emotions from Brain Signals on 3D VAD Space via Artificial Intelligence Techniques. Diagnostics. 2023; 13(13):2141. https://doi.org/10.3390/diagnostics13132141
Chicago/Turabian StyleIşık, Ümran, Ayşegül Güven, and Turgay Batbat. 2023. "Evaluation of Emotions from Brain Signals on 3D VAD Space via Artificial Intelligence Techniques" Diagnostics 13, no. 13: 2141. https://doi.org/10.3390/diagnostics13132141