I. Made Agus Wirawan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
I. Made Agus Wirawan has a total of 83 co-authors. Here are the top co-authors by number of shared publications:
I Made
Jurusan Pendidikan Teknik Informatika Universitas Pendidikan Ganesha Singaraja Bali
7 shared publications- Gede Aditra Pradnyana
Universitas Pendidikan Ganesha
5 shared publications - I Gede Partha Sindu
Universitas Pendidikan Ganesha
4 shared publications - I. Gede Mahendra Darmawiguna
Universitas Pendidikan Ganesha
4 shared publications - I Nyoman Sukajaya
Universitas Pendidikan Ganesha
3 shared publications - I. Made Gede Sunarya
Universitas Pendidikan Ganesha, Indonesia
3 shared publications
Latest Publications
- Open Access
- 18 May 2026
- Published by BON VIEW PUBLISHING PTE in Journal of Computational and Cognitive Engineering
- Vol. 5 (3) , 425-437
Optimization of Electroencephalogram Feature Based on Multi-Band Feature Matrix and Relative Difference for Imagined Motor Movement Pattern Classification
Electroencephalogram (EEG)-based imagined motor movement recognition plays a crucial role in brain–computer interface applications and medical rehabilitation systems. However, EEG signals inherently exhibit high levels of noise, are nonlinear and nonstationary, contain spatial information, and show significant signal pattern variance across individuals, complicating feature extraction and classification. This study proposes a novel framework that integrates the relative difference method with a multi-band feature matrix, explicitly incorporating four EEG frequency bands (theta, alpha, beta, and gamma) mapped according to the international system 10–20. The extracted features are processed by a three-layer convolutional neural network (CNN) for classification. The proposed method is evaluated on the MIMED dataset for six-class imagined motor movement recognition and validated on the AMIGOS dataset for four-class emotion recognition. The experimental evaluation on the MIMED dataset demonstrates consistently strong classification performance, achieving 99.36% accuracy, 99.37% precision, 99.37% recall, and an F1-score of 99.36%. These results significantly outperform those without baseline reduction, the difference, and the fractional methods. Further validation on AMIGOS shows that the proposed model achieves an accuracy of 99.64%. This result surpasses the previously reported state of the art. These findings confirm that combining relative difference and multi-band feature matrices produces more stable, representative, and discriminative EEG features, making them readily recognizable by CNN methods. The proposed approach has strong potential for real-time brain–computer interfaces, clinical neurorehabilitation, and emotion-sensitive human–computer interaction systems Received: 10 August 2025 | Revised: 14 February 2026 | Accepted: 10 April 2026 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The MIMED dataset that supports the findings of this study is openly available in Mendeley Data at https://data.mendeley.com/datasets/zs25xxjkm9/3. The AMIGOS dataset that supports the findings of this study is openly available at https://doi.org/10.1109/TAFFC.2018.2884461, reference number [18]. Author Contribution Statement I Made Agus Wirawan: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration, Funding acquisition. I Nyoman Sukajaya: Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Project administration. Ni Nyoman Mestri Agustini: Methodology, Validation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration.
- 18 May 2026
- Published by BON VIEW PUBLISHING PTE in Journal of Computational and Cognitive Engineering
- Vol. 5 (3) , 425-437
Open AccessOptimization of Electroencephalogram Feature Based on Multi-Band Feature Matrix and Relative Difference for Imagined Motor Movement Pattern Classification
Electroencephalogram (EEG)-based imagined motor movement recognition plays a crucial role in brain–computer interface applications and medical rehabilitation systems. However, EEG signals inherently exhibit high levels of noise, are nonlinear and nonstationary, contain spatial information, and show significant signal pattern variance across individuals, complicating feature extraction and classification. This study proposes a novel framework that integrates the relative difference method with a multi-band feature matrix, explicitly incorporating four EEG frequency bands (theta, alpha, beta, and gamma) mapped according to the international system 10–20. The extracted features are processed by a three-layer convolutional neural network (CNN) for classification. The proposed method is evaluated on the MIMED dataset for six-class imagined motor movement recognition and validated on the AMIGOS dataset for four-class emotion recognition. The experimental evaluation on the MIMED dataset demonstrates consistently strong classification performance, achieving 99.36% accuracy, 99.37% precision, 99.37% recall, and an F1-score of 99.36%. These results significantly outperform those without baseline reduction, the difference, and the fractional methods. Further validation on AMIGOS shows that the proposed model achieves an accuracy of 99.64%. This result surpasses the previously reported state of the art. These findings confirm that combining relative difference and multi-band feature matrices produces more stable, representative, and discriminative EEG features, making them readily recognizable by CNN methods. The proposed approach has strong potential for real-time brain–computer interfaces, clinical neurorehabilitation, and emotion-sensitive human–computer interaction systems Received: 10 August 2025 | Revised: 14 February 2026 | Accepted: 10 April 2026 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The MIMED dataset that supports the findings of this study is openly available in Mendeley Data at https://data.mendeley.com/datasets/zs25xxjkm9/3. The AMIGOS dataset that supports the findings of this study is openly available at https://doi.org/10.1109/TAFFC.2018.2884461, reference number [18]. Author Contribution Statement I Made Agus Wirawan: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration, Funding acquisition. I Nyoman Sukajaya: Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Project administration. Ni Nyoman Mestri Agustini: Methodology, Validation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration.
- 1 April 2026
- Published by Springer Nature
- p. 51-64
IONIC-Vue Based Mobile Computing System Design for Smart Pole Air Quality Monitoring in Urban Areas
- 1 April 2026
- Published by Springer Nature
- p. 51-64
IONIC-Vue Based Mobile Computing System Design for Smart Pole Air Quality Monitoring in Urban Areas
- Open Access
- 1 March 2026
- Published by Institute of Advanced Engineering and Science in International Journal of Advances in Applied Sciences
- Vol. 15 (1) , 55-64
Performance comparison of feature extraction methods for electroencephalogram-based recognition of Balinese script
Recognizing Balinese script from electroencephalogram (EEG) signals remains a challenging problem due to low signal amplitude, non-stationary dynamics, and significant inter-subject variability. Despite previous attempts, no single feature extraction method has been universally effective in addressing these limitations. To fill this gap, this study systematically evaluates five feature extraction techniques—differential entropy (DE), power spectral density (PSD), discrete wavelet transforms (DWT), Hjorth parameters, and statistical features—on the Balinese imagined spelling using electroencephalography (BISE) dataset, which contains EEG recordings specifically designed for Balinese script recognition. For classification, both artificial neural networks (ANN) and support vector machines (SVM) are applied, and their performance is validated across multiple experimental settings. Results demonstrate that DE consistently provides more stable and discriminative features than the other methods, achieving the highest classification accuracy when combined with ANN. These findings highlight the potential of DE-based approaches to advance electroencephalogram driven Balinese script recognition, offering a culturally significant contribution to brain-computer interface (BCI) research and supporting future applications in inclusive artificial intelligence, digital heritage preservation, and assistive technologies.
- 1 March 2026
- Published by Institute of Advanced Engineering and Science in International Journal of Advances in Applied Sciences
- Vol. 15 (1) , 55-64
Open AccessPerformance comparison of feature extraction methods for electroencephalogram-based recognition of Balinese script
Recognizing Balinese script from electroencephalogram (EEG) signals remains a challenging problem due to low signal amplitude, non-stationary dynamics, and significant inter-subject variability. Despite previous attempts, no single feature extraction method has been universally effective in addressing these limitations. To fill this gap, this study systematically evaluates five feature extraction techniques—differential entropy (DE), power spectral density (PSD), discrete wavelet transforms (DWT), Hjorth parameters, and statistical features—on the Balinese imagined spelling using electroencephalography (BISE) dataset, which contains EEG recordings specifically designed for Balinese script recognition. For classification, both artificial neural networks (ANN) and support vector machines (SVM) are applied, and their performance is validated across multiple experimental settings. Results demonstrate that DE consistently provides more stable and discriminative features than the other methods, achieving the highest classification accuracy when combined with ANN. These findings highlight the potential of DE-based approaches to advance electroencephalogram driven Balinese script recognition, offering a culturally significant contribution to brain-computer interface (BCI) research and supporting future applications in inclusive artificial intelligence, digital heritage preservation, and assistive technologies.
- Open Access
- 15 November 2025
- Published by Politeknik Negeri Bengkalis in INOVTEK Polbeng - Seri Informatika
- Vol. 10 (3) , 1402-1411
Improving Butterfly Fish Image Classification Accuracy using HSV Feature Extraction and SMOTE-Based Data Balancing
Class imbalance in image data can reduce the accuracy of classification models, especially when the minority class data is much smaller than the majority class. This research focuses on enhancing the classification accuracy of butterflyfish images through the application of the Synthetic Minority Over-sampling Technique (SMOTE) for data balancing, combined with the K-Nearest Neighbor (KNN) algorithm utilizing HSV-based feature extraction. The datasets were collected in two conditions, namely conditioned (controlled background and lighting) and unconditioned (varied background and natural lighting). The research stages include preprocessing, HSV feature extraction, data balancing with SMOTE, and classification using KNN with various k values (3, 5, 7, 9) and cross-validation (k-fold 5 and 10). The experimental results show that SMOTE consistently improves accuracy on both types of datasets, with the best performance at k = 3 and k-fold = 10, namely 85.32% (conditioned) and 87.59% (unconditioned). This improvement occurs because a more balanced data distribution allows the model to optimally recognize features between classes. This study proves that the integration of SMOTE and KNN is effective in overcoming class imbalance in image classification, with potential applications in the fields of digital image technology, ecosystem management, and species identification.
- 15 November 2025
- Published by Politeknik Negeri Bengkalis in INOVTEK Polbeng - Seri Informatika
- Vol. 10 (3) , 1402-1411
Open AccessImproving Butterfly Fish Image Classification Accuracy using HSV Feature Extraction and SMOTE-Based Data Balancing
Class imbalance in image data can reduce the accuracy of classification models, especially when the minority class data is much smaller than the majority class. This research focuses on enhancing the classification accuracy of butterflyfish images through the application of the Synthetic Minority Over-sampling Technique (SMOTE) for data balancing, combined with the K-Nearest Neighbor (KNN) algorithm utilizing HSV-based feature extraction. The datasets were collected in two conditions, namely conditioned (controlled background and lighting) and unconditioned (varied background and natural lighting). The research stages include preprocessing, HSV feature extraction, data balancing with SMOTE, and classification using KNN with various k values (3, 5, 7, 9) and cross-validation (k-fold 5 and 10). The experimental results show that SMOTE consistently improves accuracy on both types of datasets, with the best performance at k = 3 and k-fold = 10, namely 85.32% (conditioned) and 87.59% (unconditioned). This improvement occurs because a more balanced data distribution allows the model to optimally recognize features between classes. This study proves that the integration of SMOTE and KNN is effective in overcoming class imbalance in image classification, with potential applications in the fields of digital image technology, ecosystem management, and species identification.
- Open Access
- 12 July 2025
- Published by Politeknik Ganesha in sinkron
- Vol. 9 (3) , 1073-1083
Enhancing EEG-Based Stress Detection Using ICA, Relative Difference, and Convolutional Neural Networks
: EEG-based stress detection is crucial for early mental health monitoring, but signal quality is often degraded by artifacts and baseline variability. This study proposes an optimized preprocessing method combining Independent Component Analysis (ICA) for artifact removal and Relative Difference for baseline reduction. Using the SAM-40 EEG dataset, features were extracted with Differential Entropy and structured into a 3D EEG cube to preserve spatial-frequency information. A Convolutional Neural Network (CNN) classified stress levels into low and high categories. The proposed approach achieved 94.44% accuracy, with 100% precision for the high stress class and 81.82% recall. These results highlight the effectiveness of combining ICA and baseline reduction to enhance deep learning-based EEG signal processing for stress detection.
- 12 July 2025
- Published by Politeknik Ganesha in sinkron
- Vol. 9 (3) , 1073-1083
Open AccessEnhancing EEG-Based Stress Detection Using ICA, Relative Difference, and Convolutional Neural Networks
: EEG-based stress detection is crucial for early mental health monitoring, but signal quality is often degraded by artifacts and baseline variability. This study proposes an optimized preprocessing method combining Independent Component Analysis (ICA) for artifact removal and Relative Difference for baseline reduction. Using the SAM-40 EEG dataset, features were extracted with Differential Entropy and structured into a 3D EEG cube to preserve spatial-frequency information. A Convolutional Neural Network (CNN) classified stress levels into low and high categories. The proposed approach achieved 94.44% accuracy, with 100% precision for the high stress class and 81.82% recall. These results highlight the effectiveness of combining ICA and baseline reduction to enhance deep learning-based EEG signal processing for stress detection.
- Open Access
- 1 June 2025
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 14 (3) , 2302-2314
Hybrid method for optimizing emotion recognition models on electroencephalogram signals
Two critical factors that need to be studied in emotion recognition are the differences in electroencephalogram (EEG) signal patterns caused by participant characteristics and EEG signals with spatial information. These factors significantly affect the resulting accuracy. The model proposed in this study can consider these factors. This model consists of the modified weighted mean filter method for the basic EEG signal smoothing process, the differential entropy method for the feature extraction process, the relative difference method for the baseline reduction, the 3D cube method for feature representation, and the continuous capsule network method for the classification process. Based on testing on three public datasets, this hybrid method can overcome factors affecting emotion recognition accuracy. This statement is based on the accuracy produced by this model, which outperformed the accuracy validated in previous studies.
- 1 June 2025
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 14 (3) , 2302-2314
Open AccessHybrid method for optimizing emotion recognition models on electroencephalogram signals
Two critical factors that need to be studied in emotion recognition are the differences in electroencephalogram (EEG) signal patterns caused by participant characteristics and EEG signals with spatial information. These factors significantly affect the resulting accuracy. The model proposed in this study can consider these factors. This model consists of the modified weighted mean filter method for the basic EEG signal smoothing process, the differential entropy method for the feature extraction process, the relative difference method for the baseline reduction, the 3D cube method for feature representation, and the continuous capsule network method for the classification process. Based on testing on three public datasets, this hybrid method can overcome factors affecting emotion recognition accuracy. This statement is based on the accuracy produced by this model, which outperformed the accuracy validated in previous studies.
- Open Access
- 18 March 2025
- Published by Lembaga Publikasi Ilmiah dan Penerbitan Universitas Muhammadiyah Purwokerto in JUITA: Jurnal Informatika
- p. 37-46
Subject Independent Emotion Recognition Using Electroencephalogram Signals with Continuous Capsule Network Method
Emotions play an essential role in human reasoning. Researchers have made various efforts to improve emotion classification methods. Based on the several emotion classification methods studied in previous studies, the Continuous Capsule Network method produced the highest accuracy compared to other classification methods. This method can maintain spatial information from electroencephalogram signals so they are not reduced. However, this method has only been tested subject-dependently. Based on this study, the Continuous Capsule Network method will be applied to classify emotions in the Faculty of Engineering and Vocational Studies, Universitas Pendidikan Ganesha students. The number of participants involved in this study was 17 people (10 men and 7 women). Through six subjectindependent test scenarios, the Continuous Capsule Network method produced accuracy, precision, recall, and F1 scores of 99.31%, 99.34%, 99.20%, and 99.27, respectively. At the same time, the loss value was 0.88%. In addition, the Continuous Capsule Network method produced an average training and validation time of 401.17 seconds and an average testing time of 4.67 seconds for the six test scenarios.
- 18 March 2025
- Published by Lembaga Publikasi Ilmiah dan Penerbitan Universitas Muhammadiyah Purwokerto in JUITA: Jurnal Informatika
- p. 37-46
Open AccessSubject Independent Emotion Recognition Using Electroencephalogram Signals with Continuous Capsule Network Method
Emotions play an essential role in human reasoning. Researchers have made various efforts to improve emotion classification methods. Based on the several emotion classification methods studied in previous studies, the Continuous Capsule Network method produced the highest accuracy compared to other classification methods. This method can maintain spatial information from electroencephalogram signals so they are not reduced. However, this method has only been tested subject-dependently. Based on this study, the Continuous Capsule Network method will be applied to classify emotions in the Faculty of Engineering and Vocational Studies, Universitas Pendidikan Ganesha students. The number of participants involved in this study was 17 people (10 men and 7 women). Through six subjectindependent test scenarios, the Continuous Capsule Network method produced accuracy, precision, recall, and F1 scores of 99.31%, 99.34%, 99.20%, and 99.27, respectively. At the same time, the loss value was 0.88%. In addition, the Continuous Capsule Network method produced an average training and validation time of 401.17 seconds and an average testing time of 4.67 seconds for the six test scenarios.
- Open Access
- 10 March 2025
- Published by Elsevier in Data in Brief
- Vol. 60, 111454
Acquisition Of Balinese Imagined Spelling using Electroencephalogram (BISE) Dataset
- 10 March 2025
- Published by Elsevier in Data in Brief
- Vol. 60, 111454
Open AccessAcquisition Of Balinese Imagined Spelling using Electroencephalogram (BISE) Dataset
- Open Access
- 1 January 2025
- Published by Universitas Negeri Makassar in International Journal of Language Education
- Vol. 8 (4) , 778-794
Integrating Balinese Gamelan Music with Sensorimotor Rhythm Intervention to Improve Brain Wave Synchrony and Language Processing in Autistic Children
The main objective of this study was to explore the effectiveness of combining Balinese Gamelan music with sensorimotor rhythm intervention in improving brain wave synchronization and language processing abilities in autistic children. This study used a qualitative method with an ethnographic case study approach to explore how the integration of Balinese gamelan music and sensorimotor rhythm intervention has the potential to affect brain wave synchronization and language development in autistic children. This study in-depth examines the experiences of autistic participants in the context of a culture-based intervention, which combines elements of traditional rhythm with sensorimotor therapy in a natural and authentic environment. Participants were selected through purposive sampling technique, consisting of parents of children with autism spectrum who experience limitations in language processing, aged between 5 and 15 years, who had been selected based on parental willingness and therapist recommendations. The number of participants was estimated at 25 children, sufficient for in-depth data collection and considering the variability of responses to the intervention. This study shows that the integration of Balinese gamelan music with sensorimotor rhythm intervention has the potential to improve motor synchronization, focus ability, emotional engagement, and language processing in autistic children. This intervention was also able to create a positive atmosphere that supported children's engagement in a structured musical environment. In addition, preliminary data from EEG suggests that gamelan rhythms may help create more regular brainwave synchronization, although these findings require further research to confirm. Overall, the results of this study provide new insights into the role of traditional music in supporting sensorimotor and cognitive development in children on the autism spectrum, especially in unique cultural contexts.
- 1 January 2025
- Published by Universitas Negeri Makassar in International Journal of Language Education
- Vol. 8 (4) , 778-794
Open AccessIntegrating Balinese Gamelan Music with Sensorimotor Rhythm Intervention to Improve Brain Wave Synchrony and Language Processing in Autistic Children
The main objective of this study was to explore the effectiveness of combining Balinese Gamelan music with sensorimotor rhythm intervention in improving brain wave synchronization and language processing abilities in autistic children. This study used a qualitative method with an ethnographic case study approach to explore how the integration of Balinese gamelan music and sensorimotor rhythm intervention has the potential to affect brain wave synchronization and language development in autistic children. This study in-depth examines the experiences of autistic participants in the context of a culture-based intervention, which combines elements of traditional rhythm with sensorimotor therapy in a natural and authentic environment. Participants were selected through purposive sampling technique, consisting of parents of children with autism spectrum who experience limitations in language processing, aged between 5 and 15 years, who had been selected based on parental willingness and therapist recommendations. The number of participants was estimated at 25 children, sufficient for in-depth data collection and considering the variability of responses to the intervention. This study shows that the integration of Balinese gamelan music with sensorimotor rhythm intervention has the potential to improve motor synchronization, focus ability, emotional engagement, and language processing in autistic children. This intervention was also able to create a positive atmosphere that supported children's engagement in a structured musical environment. In addition, preliminary data from EEG suggests that gamelan rhythms may help create more regular brainwave synchronization, although these findings require further research to confirm. Overall, the results of this study provide new insights into the role of traditional music in supporting sensorimotor and cognitive development in children on the autism spectrum, especially in unique cultural contexts.
- Open Access
- 1 December 2024
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 13 (4) , 4263-4272
Improving baseline reduction for emotion recognition based on electroencephalogram signals
The baseline reduction method has been widely used to define electroencephalogram (EEG) signal patterns. However, because the baseline signal in this approach contains artifacts, the baseline reduction approach cannot perform optimally. As a result, decreasing artifacts in the baseline signal is critical. The mean, Gaussian, and Savitzky-Golay filters will be compared in this study to minimize artifacts in the baseline signal. Three secondary datasets are utilized to evaluate these approaches' capacity to remove artifacts. These three strategies are also tested with the convolution neural network classification algorithms. When applied to the dataset for emotion analysis using physiological signals (DEAP) and a dataset for multimodal research of affect, personality traits, and mood on individuals and groups (AMIGOS) datasets, the mean filter can increase baseline reduction performance based on twenty-four test scenarios. On the data readiness for machine learning research (DREAMER) dataset, however, the Gaussian filter is preferable. The relative difference approach was employed in this study's baseline reduction process to generate EEG signal patterns that are easy to recognize throughout the classification phase, which impacts increasing accuracy.
- 1 December 2024
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 13 (4) , 4263-4272
Open AccessImproving baseline reduction for emotion recognition based on electroencephalogram signals
The baseline reduction method has been widely used to define electroencephalogram (EEG) signal patterns. However, because the baseline signal in this approach contains artifacts, the baseline reduction approach cannot perform optimally. As a result, decreasing artifacts in the baseline signal is critical. The mean, Gaussian, and Savitzky-Golay filters will be compared in this study to minimize artifacts in the baseline signal. Three secondary datasets are utilized to evaluate these approaches' capacity to remove artifacts. These three strategies are also tested with the convolution neural network classification algorithms. When applied to the dataset for emotion analysis using physiological signals (DEAP) and a dataset for multimodal research of affect, personality traits, and mood on individuals and groups (AMIGOS) datasets, the mean filter can increase baseline reduction performance based on twenty-four test scenarios. On the data readiness for machine learning research (DREAMER) dataset, however, the Gaussian filter is preferable. The relative difference approach was employed in this study's baseline reduction process to generate EEG signal patterns that are easy to recognize throughout the classification phase, which impacts increasing accuracy.