Mahmud Dwi Sulistiyo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Mahmud Dwi Sulistiyo has a total of 61 co-authors. Here are the top co-authors by number of shared publications:
- Aditya Firman Ihsan
Telkom University, Bandung, Indonesia
2 shared publications - Adiwijaya Adiwijaya
Telkom University
2 shared publications - Ema Rachmawati
Telkom University
2 shared publications - Syafrial Fachri Pane
Universitas Logistik dan Bisnis Internasional
2 shared publications - Aditya Firman Ihsan
School of Computing, Telkom University
1 shared publication - Aji Gautama Putrada
Telkom University
1 shared publication
Latest Publications
- 2 August 2026
- Published by Springer Nature
- p. 531-541
Energy-Efficient Image Compression for Business Intelligence: Comparative Insights on Wavelet-Based Methods
- 2 August 2026
- Published by Springer Nature
- p. 531-541
Energy-Efficient Image Compression for Business Intelligence: Comparative Insights on Wavelet-Based Methods
- Open Access
- 1 April 2026
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 15  (2)
MNetNCR: MobileNet model for efficient traditional Nusantara script character recognition
Preservation of traditional Nusantara scripts is very important because these traditional scripts are part of the cultural heritage that reflects the identity and history of the nation. This research proposed MobileNet for Nusantara character recognition (MNetNCR) model based on MobileNetV3 architecture to recognize traditional Nusantara scripts with lightweight, efficient architecture, and accurate recognition. The novel and comprehensive datasets for traditional Nusantara scripts have been curated in this research, that will later be stored digitally and can be used in further research. This novel dataset includes handwritten Balinese, Batak, Javanese, Lontara, and Sundanese scripts, each with unique visual characteristics. The proposed MNetNCR model is highly effective in recognizing characters, achieving F1-scores of 0.9934 for Balinese, 0.9450 for Batak, 0.9788 for Javanese, 0.9936 for Lontara, and 0.9961 for Sundanese scripts, according to the experimental results. The MNetNCR model built in this research has been proven to be effective and efficient in recognizing traditional scripts accurately. It also supports the preservation and promotion of the nation's cultural and historical heritage.
- 1 April 2026
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 15  (2)
Open AccessMNetNCR: MobileNet model for efficient traditional Nusantara script character recognition
Preservation of traditional Nusantara scripts is very important because these traditional scripts are part of the cultural heritage that reflects the identity and history of the nation. This research proposed MobileNet for Nusantara character recognition (MNetNCR) model based on MobileNetV3 architecture to recognize traditional Nusantara scripts with lightweight, efficient architecture, and accurate recognition. The novel and comprehensive datasets for traditional Nusantara scripts have been curated in this research, that will later be stored digitally and can be used in further research. This novel dataset includes handwritten Balinese, Batak, Javanese, Lontara, and Sundanese scripts, each with unique visual characteristics. The proposed MNetNCR model is highly effective in recognizing characters, achieving F1-scores of 0.9934 for Balinese, 0.9450 for Batak, 0.9788 for Javanese, 0.9936 for Lontara, and 0.9961 for Sundanese scripts, according to the experimental results. The MNetNCR model built in this research has been proven to be effective and efficient in recognizing traditional scripts accurately. It also supports the preservation and promotion of the nation's cultural and historical heritage.
- Open Access
- 29 November 2025
- Published by Ikatan Ahli Informatika Indonesia (IAII) in Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
- Vol. 9  (6) , 1538-1550
Investigating Shallow Learning Methods for Optical Character Recognition of Indonesia’s Nusantara Scripts
Indonesia has numerous regional scripts—or so-called Nusantara scripts—and recognizing them is important to preserve Indonesia's cultural heritage. The advances of AI and computer vision technologies make it possible for a machine to optically read the handwritten scripts through the Optical Character Recognition (OCR) technique. However, collecting some of the top OCR solutions and comprehensively investigating their performances on the Nusantara scripts is currently lacking. This study investigates and evaluates some shallow learning-based methods on our newly introduced datasets, consisting of more than 38,000-character images across 80 letter classes in total; here, we focus on three regional scripts: Javanese, Sundanese, and Balinese. The methods include Random Forest, SVM, Logistic Regression, and Gaussian Naïve Bayes, as well as boosting techniques such as XGBoost, Light GBM, and CatBoost. A 5-fold cross-validation approach assessed model performance based on accuracy, precision, recall, and F1-score. Based on the experimental results, the methods demonstrated their competitiveness in reaching the best models for scripts; in particular, XGBoost, Light GBM, and Random Forest-Gini were the winners for Javanese, Sundanese, and Balinese scripts, respectively. These findings demonstrate the effectiveness of ensemble learning methods for diverse handwritten scripts. Comparative analysis to prior deep learning studies is also discussed in this paper. In addition, this research also contributes to preserving Indonesian traditional scripts, as well as offers insights for future regional OCR in other countries.
- 29 November 2025
- Published by Ikatan Ahli Informatika Indonesia (IAII) in Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
- Vol. 9  (6) , 1538-1550
Open AccessInvestigating Shallow Learning Methods for Optical Character Recognition of Indonesia’s Nusantara Scripts
Indonesia has numerous regional scripts—or so-called Nusantara scripts—and recognizing them is important to preserve Indonesia's cultural heritage. The advances of AI and computer vision technologies make it possible for a machine to optically read the handwritten scripts through the Optical Character Recognition (OCR) technique. However, collecting some of the top OCR solutions and comprehensively investigating their performances on the Nusantara scripts is currently lacking. This study investigates and evaluates some shallow learning-based methods on our newly introduced datasets, consisting of more than 38,000-character images across 80 letter classes in total; here, we focus on three regional scripts: Javanese, Sundanese, and Balinese. The methods include Random Forest, SVM, Logistic Regression, and Gaussian Naïve Bayes, as well as boosting techniques such as XGBoost, Light GBM, and CatBoost. A 5-fold cross-validation approach assessed model performance based on accuracy, precision, recall, and F1-score. Based on the experimental results, the methods demonstrated their competitiveness in reaching the best models for scripts; in particular, XGBoost, Light GBM, and Random Forest-Gini were the winners for Javanese, Sundanese, and Balinese scripts, respectively. These findings demonstrate the effectiveness of ensemble learning methods for diverse handwritten scripts. Comparative analysis to prior deep learning studies is also discussed in this paper. In addition, this research also contributes to preserving Indonesian traditional scripts, as well as offers insights for future regional OCR in other countries.
- Open Access
- 17 November 2025
- Published by Universitas Ahmad Dahlan in International Journal of Advances in Intelligent Informatics
- Vol. 11  (4) , 653-666
PSO-Enhanced ensemble techniques for pandemic prediction and feature importance analysis
During the pandemic crisis that hit after 2020, Indonesia, like many other countries, faced tremendous challenges in areas such as health, economy, and mobility. An in-depth understanding of the dynamics and changes in these areas is essential to address the impacts of the pandemic. This research is an attempt to deeply analyze the impact of the pandemic and the most effective forecasting methods based on data and phenomena. Indonesia, with its growing economy and constantly adapting health system, faces conventional economic impacts, while its health system response tries to keep up with urgent needs driven by the spread of the virus. In the context of mobility, changes in how people move and interact significantly affect virus transmission. Modeling a pandemic event with all its complexities is not an easy task. Even more so, in finding the right method for prediction, ensemble techniques such as stacking and regression voting are emerging as promising approaches. However, deep learning and particle swarm optimization (PSO) techniques offer new innovations. The results of this study show that the ensemble vote provides the best performance in predicting confirmed positive cases and mortality based on factors of health, economic and population mobility in Indonesia. Through feature importance analysis using MDI and Tree SHAP, we conclude that factors such as active cases, the number of vaccinations, and economic indicators, such as close IDR and close IHSG, have a significant influence on the growth of confirmed positive cases. Meanwhile, recovery factors and vaccination number play an important role in the growth of the number of death cases. This study confirms that a multivariate approach that considers health, economy and mobility is the key to understanding and responding more effectively to the pandemic in Indonesia.
- 17 November 2025
- Published by Universitas Ahmad Dahlan in International Journal of Advances in Intelligent Informatics
- Vol. 11  (4) , 653-666
Open AccessPSO-Enhanced ensemble techniques for pandemic prediction and feature importance analysis
During the pandemic crisis that hit after 2020, Indonesia, like many other countries, faced tremendous challenges in areas such as health, economy, and mobility. An in-depth understanding of the dynamics and changes in these areas is essential to address the impacts of the pandemic. This research is an attempt to deeply analyze the impact of the pandemic and the most effective forecasting methods based on data and phenomena. Indonesia, with its growing economy and constantly adapting health system, faces conventional economic impacts, while its health system response tries to keep up with urgent needs driven by the spread of the virus. In the context of mobility, changes in how people move and interact significantly affect virus transmission. Modeling a pandemic event with all its complexities is not an easy task. Even more so, in finding the right method for prediction, ensemble techniques such as stacking and regression voting are emerging as promising approaches. However, deep learning and particle swarm optimization (PSO) techniques offer new innovations. The results of this study show that the ensemble vote provides the best performance in predicting confirmed positive cases and mortality based on factors of health, economic and population mobility in Indonesia. Through feature importance analysis using MDI and Tree SHAP, we conclude that factors such as active cases, the number of vaccinations, and economic indicators, such as close IDR and close IHSG, have a significant influence on the growth of confirmed positive cases. Meanwhile, recovery factors and vaccination number play an important role in the growth of the number of death cases. This study confirms that a multivariate approach that considers health, economy and mobility is the key to understanding and responding more effectively to the pandemic in Indonesia.
- Open Access
- 2 September 2025
- Published by Fakultas Sains dan Teknologi, Unira Malang in I-Com: Indonesian Community Journal
- Vol. 5  (3) , 1320-1329
Menjembatani Teknologi dan Spiritualitas: Pengenalan Artificial Intelligence di Pondok Pesantren melalui Workshop Kolaboratif
Pesantren sebagai lembaga pendidikan berbasis keagamaan memiliki tantangan dalam mengakses perkembangan teknologi terbaru seperti Artificial Intelligence (AI). Menjawab kebutuhan ini, tim pengabdian dari Telkom University menyelenggarakan workshop pengenalan AI dan mengundang sivitas Pondok Pesantren Modern Assuruur. Kegiatan ini bertujuan untuk meningkatkan literasi digital para santri melalui pengenalan konsep dasar AI dan pemanfaatannya dalam pembelajaran Al-Qur’an. Workshop dilakukan secara hybrid bersama narasumber dari Telkom University dan Universiti Putra Malaysia (UPM), mencakup pretest, materi inti, praktik aplikasi Tarteel, dan posttest. Hasil evaluasi menunjukkan peningkatan skor pemahaman peserta dari rata-rata 93,97 menjadi 98,10, disertai penurunan standar deviasi dari 14,50 menjadi 5,88. Kegiatan ini membuktikan bahwa pendekatan yang kontekstual dan aplikatif mampu meningkatkan pemahaman santri terhadap teknologi modern serta membuka peluang integrasi AI dalam pendidikan Islam.
- 2 September 2025
- Published by Fakultas Sains dan Teknologi, Unira Malang in I-Com: Indonesian Community Journal
- Vol. 5  (3) , 1320-1329
Open AccessMenjembatani Teknologi dan Spiritualitas: Pengenalan Artificial Intelligence di Pondok Pesantren melalui Workshop Kolaboratif
Pesantren sebagai lembaga pendidikan berbasis keagamaan memiliki tantangan dalam mengakses perkembangan teknologi terbaru seperti Artificial Intelligence (AI). Menjawab kebutuhan ini, tim pengabdian dari Telkom University menyelenggarakan workshop pengenalan AI dan mengundang sivitas Pondok Pesantren Modern Assuruur. Kegiatan ini bertujuan untuk meningkatkan literasi digital para santri melalui pengenalan konsep dasar AI dan pemanfaatannya dalam pembelajaran Al-Qur’an. Workshop dilakukan secara hybrid bersama narasumber dari Telkom University dan Universiti Putra Malaysia (UPM), mencakup pretest, materi inti, praktik aplikasi Tarteel, dan posttest. Hasil evaluasi menunjukkan peningkatan skor pemahaman peserta dari rata-rata 93,97 menjadi 98,10, disertai penurunan standar deviasi dari 14,50 menjadi 5,88. Kegiatan ini membuktikan bahwa pendekatan yang kontekstual dan aplikatif mampu meningkatkan pemahaman santri terhadap teknologi modern serta membuka peluang integrasi AI dalam pendidikan Islam.
- Open Access
- 1 September 2025
- Published by Bright Publisher in Journal of Applied Data Sciences
- Vol. 6  (4) , 2938-2958
Hybrid Multi-Objective Metaheuristic Machine Learning for Optimizing Pandemic Growth Prediction
Hybrid Multi-Objective Metaheuristic Machine Learning for Optimizing Pandemic Growth Prediction
- 1 September 2025
- Published by Bright Publisher in Journal of Applied Data Sciences
- Vol. 6  (4) , 2938-2958
Open AccessHybrid Multi-Objective Metaheuristic Machine Learning for Optimizing Pandemic Growth Prediction
Hybrid Multi-Objective Metaheuristic Machine Learning for Optimizing Pandemic Growth Prediction
- Open Access
- 11 July 2025
- Published by Ikatan Ahli Informatika Indonesia (IAII) in Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
- Vol. 9  (4) , 796-804
Advancing Vehicle Logo Detection with DETR to Handle Small Logos and Low-Quality Images
Image-based vehicle logo detection is an important component in the implementation of vehicle information recognition technology, which supports the development of intelligent transportation systems. Vehicle logos, as elements that represent the identities of vehicle brands and models, play a significant role in completing vehicle identity data. The information obtained from this logo can be utilized to solve various traffic problems, such as vehicle document counterfeiting and theft, and for better traffic planning and management purposes. However, the main challenge in developing an accurate logo detection system lies in the wide variety of shapes, sizes, and positions of logos in different types of vehicles. In addition, the generally small size of logos, especially on certain vehicles, often makes it difficult for computer-based detection systems to recognize logos consistently, thus affecting the overall performance of the detection model. In this research, the Detection Transformers (DETR) method is used to build a vehicle logo detection system that focuses on small-scale logo. The testing process was conducted using the VL-10 dataset, which was specifically designed for vehicle logo detection evaluation. The results show that the DETR model can detect vehicle logos very well, even for small-scale logos. The model achieved an AP50 value of 0.952, which indicates a high level of accuracy and reliability in detecting the vehicle logo in the dataset used.
- 11 July 2025
- Published by Ikatan Ahli Informatika Indonesia (IAII) in Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
- Vol. 9  (4) , 796-804
Open AccessAdvancing Vehicle Logo Detection with DETR to Handle Small Logos and Low-Quality Images
Image-based vehicle logo detection is an important component in the implementation of vehicle information recognition technology, which supports the development of intelligent transportation systems. Vehicle logos, as elements that represent the identities of vehicle brands and models, play a significant role in completing vehicle identity data. The information obtained from this logo can be utilized to solve various traffic problems, such as vehicle document counterfeiting and theft, and for better traffic planning and management purposes. However, the main challenge in developing an accurate logo detection system lies in the wide variety of shapes, sizes, and positions of logos in different types of vehicles. In addition, the generally small size of logos, especially on certain vehicles, often makes it difficult for computer-based detection systems to recognize logos consistently, thus affecting the overall performance of the detection model. In this research, the Detection Transformers (DETR) method is used to build a vehicle logo detection system that focuses on small-scale logo. The testing process was conducted using the VL-10 dataset, which was specifically designed for vehicle logo detection evaluation. The results show that the DETR model can detect vehicle logos very well, even for small-scale logos. The model achieved an AP50 value of 0.952, which indicates a high level of accuracy and reliability in detecting the vehicle logo in the dataset used.
- Open Access
- 15 December 2024
- Published by Yayasan Riset dan Pengembangan Intelektual in Journal of Applied Engineering and Technological Science (JAETS)
- Vol. 6  (1) , 561-578
A Combined MobileNetV2 and CBAM Model to Improve Classifying the Breast Cancer Ultrasound Images
Breast cancer is the main cause of death in women throughout the world. Early detection using ultrasound is very necessary to reduce cases of breast cancer. However, the ultrasound analysis process requires a lot of time and medical personnel because classification is difficult due to noise, complex texture, and subjective assessment. Previous studies were successful in ultrasound classification of breast cancer but required large computations and complex models. This research aims to overcome these shortcomings by using a lighter but more accurate model. We integrated the CBAM attention module into the MobileNetV2 model to improve breast cancer detection accuracy, speed up diagnosis, and reduce computational requirements. Gradient Weighted Class Activation Mapping (Grad-CAM) is used to improve classification explanations. Ultrasound images from two databases were combined to train, validate, and test this model. The test results show that MobileNetV2-CBAM achieves a test accuracy of 93%, higher than the complex models VGG-16 (80%), VGG-19 (82%), InceptionV3 (80%), and ResNet-50 (84%). CBAM is proven to improve MobileNetV2 performance with an 11% increase in accuracy. Grad-CAM visualization shows that MobileNetV2-CBAM can better focus on localizing important regions in breast cancer images, providing clearer explanations and assisting medical personnel in diagnosis.
- 15 December 2024
- Published by Yayasan Riset dan Pengembangan Intelektual in Journal of Applied Engineering and Technological Science (JAETS)
- Vol. 6  (1) , 561-578
Open AccessA Combined MobileNetV2 and CBAM Model to Improve Classifying the Breast Cancer Ultrasound Images
Breast cancer is the main cause of death in women throughout the world. Early detection using ultrasound is very necessary to reduce cases of breast cancer. However, the ultrasound analysis process requires a lot of time and medical personnel because classification is difficult due to noise, complex texture, and subjective assessment. Previous studies were successful in ultrasound classification of breast cancer but required large computations and complex models. This research aims to overcome these shortcomings by using a lighter but more accurate model. We integrated the CBAM attention module into the MobileNetV2 model to improve breast cancer detection accuracy, speed up diagnosis, and reduce computational requirements. Gradient Weighted Class Activation Mapping (Grad-CAM) is used to improve classification explanations. Ultrasound images from two databases were combined to train, validate, and test this model. The test results show that MobileNetV2-CBAM achieves a test accuracy of 93%, higher than the complex models VGG-16 (80%), VGG-19 (82%), InceptionV3 (80%), and ResNet-50 (84%). CBAM is proven to improve MobileNetV2 performance with an 11% increase in accuracy. Grad-CAM visualization shows that MobileNetV2-CBAM can better focus on localizing important regions in breast cancer images, providing clearer explanations and assisting medical personnel in diagnosis.
- Open Access
- 11 April 2024
- Published by Springer Nature in International Journal of Computational Intelligence Systems
- Vol. 17  (1) , 1-18
Leveraging Model Scaling and Butterfly Network in the Bone Scan Image Segmentation
As we all know, cancer is one of the leading causes of death worldwide and the second leading cause of death overall. This is why regular screenings or health checks are necessary to detect cancer lesions early. Since bone scan images have become the primary means of detecting the emergence of cancer lesions on bone, high segmentation accuracy is essential for establishing the model of some predefined regions in bone scan images where cancer metastasis was predicted to appear. Consequently, robust localization and identification of the specific region in bone scan images are required for automated metastasis detection. To this end, we propose Efficient-BtrflyNet, a new deep learning-based architecture for skeleton segmentation of whole-body bone scan images. The proposed architecture exploits the benefits of EfficientNet’s model scaling and the encoder–decoder design of butterfly-type networks. We added EfficientNetB7 to the encoder section to obtain more specific features. The proposed architecture simultaneously processes anterior and posterior whole-body bone scan images. Using 37 bone scan images, we evaluated the performance of our proposed skeleton segmentation system using the Dice score. Efficient-BtrflyNet achieves superior segmentation performance compared to the existing representative method.
- 11 April 2024
- Published by Springer Nature in International Journal of Computational Intelligence Systems
- Vol. 17  (1) , 1-18
Open AccessLeveraging Model Scaling and Butterfly Network in the Bone Scan Image Segmentation
As we all know, cancer is one of the leading causes of death worldwide and the second leading cause of death overall. This is why regular screenings or health checks are necessary to detect cancer lesions early. Since bone scan images have become the primary means of detecting the emergence of cancer lesions on bone, high segmentation accuracy is essential for establishing the model of some predefined regions in bone scan images where cancer metastasis was predicted to appear. Consequently, robust localization and identification of the specific region in bone scan images are required for automated metastasis detection. To this end, we propose Efficient-BtrflyNet, a new deep learning-based architecture for skeleton segmentation of whole-body bone scan images. The proposed architecture exploits the benefits of EfficientNet’s model scaling and the encoder–decoder design of butterfly-type networks. We added EfficientNetB7 to the encoder section to obtain more specific features. The proposed architecture simultaneously processes anterior and posterior whole-body bone scan images. Using 37 bone scan images, we evaluated the performance of our proposed skeleton segmentation system using the Dice score. Efficient-BtrflyNet achieves superior segmentation performance compared to the existing representative method.