Emotion Classification Using Optimized Features and Ensemble Learning Techniques for EEG Dataset DOI

S. Dhivya Bharkavi,

S. Kavitha, M. Harini

et al.

Published: Nov. 15, 2023

Emotion recognition from electroencephalogram (EEG) signals is one of the important real time applications in Brain-Computer Interface (BCI). The proposed research addresses challenges emotion classification and analyses different ensemble learning algorithms for classifying emotions, specifically positive, negative, neutral states, EEG data. dataset used this work collected Kaggle, has 2132 samples with 2549 features, where each sample corresponds to recorded during various emotional states. To enhance performance, diverse are applied, including Random Forest, LightGBM, Extra Tree, Gradient XGBoost, AdaBoost Bagged Decision Tree. In addition, optimal set features selected through Pearson Correlation Coefficient (PCC) based on threshold value Mutual Information (MI) mi-score. Using feature set, models generated validated using suitable quantitative metrics such as accuracy, sensitivity specificity test set. From results it observed that Boosting Tree achieved above 99% LightGBM exhibited superior performance 99.81% accuracy. This outcome proves effectiveness obtained both PCC MI techniques.

Language: Английский

Electroencephalogram-Based Emotion Recognition: A Comparative Analysis of Supervised Machine Learning Algorithms DOI Creative Commons
A. H. Prakash,

Alwin Poulose

Data Science and Management, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

Language: Английский

Citations

0
A Resource-Efficient Multi-Entropy Fusion Method and Its Application for EEG-Based Emotion Recognition DOI Creative Commons
Jiawen Li, Guanyuan Feng, Ling Chen

et al.

Entropy, Journal Year: 2025, Volume and Issue: 27(1), P. 96 - 96

Published: Jan. 20, 2025

Emotion recognition is an advanced technology for understanding human behavior and psychological states, with extensive applications mental health monitoring, human–computer interaction, affective computing. Based on electroencephalography (EEG), the biomedical signals naturally generated by brain, this work proposes a resource-efficient multi-entropy fusion method classifying emotional states. First, Discrete Wavelet Transform (DWT) applied to extract five brain rhythms, i.e., delta, theta, alpha, beta, gamma, from EEG signals, followed acquisition of features, including Spectral Entropy (PSDE), Singular Spectrum (SSE), Sample (SE), Fuzzy (FE), Approximation (AE), Permutation (PE). Then, such entropies are fused into matrix represent complex dynamic characteristics EEG, denoted as Brain Rhythm Matrix (BREM). Next, Dynamic Time Warping (DTW), Mutual Information (MI), Spearman Correlation Coefficient (SCC), Jaccard Similarity (JSC) measure similarity between unknown testing BREM data positive/negative samples classification. Experiments were conducted using DEAP dataset, aiming find suitable scheme regarding measures, time windows, input numbers channel data. The results reveal that DTW yields best performance in measures 5 s window. In addition, single-channel mode outperforms single-region mode. proposed achieves 84.62% 82.48% accuracy arousal valence classification tasks, respectively, indicating its effectiveness reducing dimensionality computational complexity while maintaining over 80%. Such performances remarkable when considering limited resources concern, which opens possibilities innovative entropy can help design portable EEG-based emotion-aware devices daily usage.

Language: Английский

Citations

0
Electroencephalogram Based Emotion Recognition Using Hybrid Intelligent Method and Discrete Wavelet Transform DOI Creative Commons
Duy Nguyen, M.T. Nguyen, Kou Yamada

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(5), P. 2328 - 2328

Published: Feb. 21, 2025

Electroencephalography-based emotion recognition is essential for brain-computer interface combined with artificial intelligence. This paper proposes a novel algorithm human detection using hybrid paradigm of convolutional neural networks and boosting model. The proposed employs two subsets 18 14 features extracted from four sub-bands discrete wavelet transform. These are identified as the optimal most relevant, among 42 original input 8 6 productive channels dual genetic wise-subject 5-fold cross validation procedure in which first second algorithms address efficient feature subsets. estimated by differently intelligent models on set. produces an accuracy 70.43%/76.05%, precision 69.88%/74.57%, recall 98.70%/99.17%, F1 score 81.83%/85.13% valence/arousal classifications, suggest that frontal left regions cortex associate especially to emotions.

Language: Английский

Citations

0
Sex-Specific Ensemble Models for Type 2 Diabetes Classification in the Mexican Population DOI Creative Commons

Miguel M. Mendoza-Mendoza,

Samara Acosta-Jiménez, Carlos E. Galván-Tejada

et al.

Diabetes Metabolic Syndrome and Obesity, Journal Year: 2025, Volume and Issue: Volume 18, P. 1501 - 1525

Published: May 1, 2025

Type 2 diabetes (T2D) is considered a global pandemic by the World Health Organization (WHO), with growing prevalence, particularly in Mexico. Accurate early diagnosis remains challenge, especially when accounting for biological sex-based differences. This study aims to enhance classification of T2D Mexican population applying sex-specific ensemble models combined genetic algorithm-based feature selection. A dataset 1787 patients (895 females, 892 males) analyzed. Data are split sex, and selection performed using GALGO, tool. Classification including Random Forest, K-Nearest Neighbor, Support Vector Machine, Logistic Regression trained evaluated. Ensemble stacking constructed separately each sex improve performance. The male-specific model achieved 94% specificity 96% sensitivity, while female-specific reached 90% sensitivity. Both demonstrated strong overall proposed represent clinically valuable approach personalized diagnosis. By identifying predictive features, this work supports development precision medicine tools tailored population. contributes improving diagnostic supporting more equitable approaches clinical settings.

Language: Английский

Citations

0
Characterisation of Cognitive Load Using Machine Learning Classifiers of Electroencephalogram Data DOI Creative Commons
Qi Wang,

Daniel Smythe,

Jun Cao

et al.

Sensors, Journal Year: 2023, Volume and Issue: 23(20), P. 8528 - 8528

Published: Oct. 17, 2023

A high cognitive load can overload a person, potentially resulting in catastrophic accidents. It is therefore important to ensure the level of associated with safety-critical tasks (such as driving vehicle) remains manageable for drivers, enabling them respond appropriately changes environment. Although electroencephalography (EEG) has attracted significant interest research, few studies have used EEG investigate context driving. This paper presents feasibility study on simulation various levels through designing and implementing four tasks. We employ machine learning-based classification techniques using recordings differentiate conditions. An dataset containing these from group 20 participants was collected whether be an indicator load. The train Deep Neural Networks Support Vector Machine models. results showed that best model achieved accuracy 90.37%, utilising statistical features multiple frequency bands 24 channels. Furthermore, Gamma Beta higher than Alpha Theta during analysis. outcomes this potential enhance Human–Machine Interface vehicles, contributing improved safety.

Language: Английский

Citations

6
Artificial intelligence-based smart devices for biomedical applications DOI
Deblu Sahu, Bala Chakravarthy Neelapu,

J. Sivaraman

et al.

Elsevier eBooks, Journal Year: 2024, Volume and Issue: unknown, P. 339 - 357

Published: Jan. 1, 2024

Language: Английский

Citations

0
Application of Artificial Intelligence Methods in Processing of Emotions, Decisions, and Opinions DOI Creative Commons
Michał Ptaszyński, Paweł Dybała, Rafał Rzepka

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(13), P. 5912 - 5912

Published: July 6, 2024

The rapid advancement of artificial intelligence (AI) and natural language processing (NLP) has profoundly impacted our understanding emotions, decision-making, opinions, particularly within the context Internet social media [...]

Language: Английский

Citations

0
EEG Based Emotion Detection by Using Modified Tunicate Swarm Optimization Algorithm DOI Creative Commons
Amrendra Tripathi, Tanupriya Choudhury, Hitesh Kumar Sharma

et al.

Ingénierie des systèmes d information, Journal Year: 2024, Volume and Issue: 29(4), P. 1333 - 1342

Published: Aug. 21, 2024

In recent years, the rapid development of computer applications for automatic classification human emotions-based Electroencephalography (EEG) has significant attention from researchers.However, existing techniques have not adequately addressed contextinformation inherent in EEG signals.To address issue, this research utilized an automated model enhancing EEG-based emotion recognition.The Modified Tunicate Swarm Optimization Algorithm (MTSOA) improves recognition by context information management.It signal processing, resulting more accurate emotional state detection.This overcomes fundamental difficulties and algorithm efficacy extracting relevant data signals robust detection systems.MTSOA is used feature selection because its capacity to navigate complex search spaces effectively.Because effectively explore parameter spaces, Rat (RSOA) choose hyperparameters.According results suggested method better outcomes arousal 89.58%, valence 92.29% which was significantly higher than ensemble median empirical mode decomposition (MEEMD), CNN with SVM, Kernel matrix+DNN methods.

Language: Английский

Citations

0
A comprehensive survey of evolutionary algorithms and metaheuristics in brain EEG-based applications DOI Creative Commons
Muhammad Arif, Faizan Ur Rehman, Lukáš Sekanina

et al.

Journal of Neural Engineering, Journal Year: 2024, Volume and Issue: 21(5), P. 051002 - 051002

Published: Sept. 25, 2024

Abstract Electroencephalography (EEG) has emerged as a primary non-invasive and mobile modality for understanding the complex workings of human brain, providing invaluable insights into cognitive processes, neurological disorders, brain–computer interfaces. Nevertheless, volume EEG data, presence artifacts, selection optimal channels, need feature extraction from data present considerable challenges in achieving meaningful distinguishing outcomes machine learning algorithms utilized to process data. Consequently, demand sophisticated optimization techniques become imperative overcome these hurdles effectively. Evolutionary (EAs) other nature-inspired metaheuristics have been applied powerful design tools recent years, showcasing their significance addressing various problems relevant brain EEG-based applications. This paper presents comprehensive survey highlighting importance EAs The is organized according main areas where applied, namely artifact mitigation, channel selection, extraction, signal classification. Finally, current future aspects context applications are discussed.

Language: Английский

Citations

0
Detection of ovarian cancer using a methodology with feature extraction and selection with genetic algorithms and machine learning DOI
Samara Acosta-Jiménez,

Miguel M. Mendoza-Mendoza,

Carlos E. Galván-Tejada

et al.

Network Modeling Analysis in Health Informatics and Bioinformatics, Journal Year: 2024, Volume and Issue: 14(1)

Published: Dec. 19, 2024

Language: Английский

Citations

0