Aslan, Emrah
Loading...
Profile URL
Name Variants
Aslan, E.
Aslan, Emrah
Aslan, Emrah
Job Title
Dr. Öğr. Üyesi
Email Address
Main Affiliation
Department of Computer Engineering / Bilgisayar Mühendisliği Bölümü
Status
Current Staff
Website
ORCID ID
Scopus Author ID
Turkish CoHE Profile ID
Google Scholar ID
WoS Researcher ID
Sustainable Development Goals
2
ZERO HUNGER
2
Research Products
3
GOOD HEALTH AND WELL-BEING
4
Research Products
7
AFFORDABLE AND CLEAN ENERGY
3
Research Products
8
DECENT WORK AND ECONOMIC GROWTH
2
Research Products
11
SUSTAINABLE CITIES AND COMMUNITIES
1
Research Products
14
LIFE BELOW WATER
1
Research Products
This researcher does not have a Scopus ID.
This researcher does not have a WoS ID.
Scholarly Output
14
Articles
14
Views / Downloads
78/0
Supervised MSc Theses
0
Supervised PhD Theses
0
WoS Citation Count
64
Scopus Citation Count
90
WoS h-index
4
Scopus h-index
5
Patents
0
Projects
0
WoS Citations per Publication
4.57
Scopus Citations per Publication
6.43
Open Access Source
8
Supervised Theses
0
Google Analytics Visitor Traffic
| Journal | Count |
|---|---|
| IEEE Access | 2 |
| Chemometrics and Intelligent Laboratory Systems | 1 |
| Gazi University Journal of Science Part A: Engineering and Innovation | 1 |
| International Journal of Integrated Engineering | 1 |
| Journal of Applied Science and Technology Trends | 1 |
Current Page: 1 / 3
Scopus Quartile Distribution
Competency Cloud
14 results
Scholarly Output Search Results
Now showing 1 - 10 of 14
Article Citation - WoS: 9Citation - Scopus: 10A Hybrid Machine Learning Approach for Predicting Power Transformer Failures Using Internet of Things-Based Monitoring and Explainable Artificial Intelligence(IEEE - Inst Electrical Electronics Engineers inc, 2025) Ozupak, Yildirim; Alpsalaz, Feyyaz; Elbarbary, Zakaria M. S.; Aslan, EmrahPower transformers are critical components in ensuring the continuous and stable operation of power systems. Failures in these units can lead to significant power outages and costly downtime. Existing maintenance strategies often fail to accurately predict such failures, highlighting the need for novel predictive approaches. This study aims to improve the reliability of power systems by predicting transformer failures through the integration of IoT technologies and advanced machine learning techniques. The proposed hybrid model combines the LightGBM algorithm with GridSearch optimization to achieve both high predictive accuracy and computational efficiency. In addition, the model enhances interpretability by incorporating SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) for transparent decision making. The study presents a detailed comparison of different classification algorithms and evaluates their performance using metrics such as accuracy, recall, and F1 score. The results show that the hybrid model outperforms other methods, achieving an accuracy of 99.91%. The SHAP and LIME analyses provide engineers and researchers with valuable insights by highlighting the most influential features in failure prediction. In addition, the model's ability to efficiently handle large data sets enhances its practicality in real-world power systems. By proposing an innovative approach to failure prediction, this research contributes to both the theoretical foundation and practical advancement of sustainable and reliable energy infrastructures.Article Citation - WoS: 2Citation - Scopus: 2Advanced Fault Classification in Induction Motors for Electric Vehicles Using a Stacking Ensemble Learning Approach(MDPI, 2025) Benkaihoul, Said; Khadar, Saad; Ozupak, Yildirim; Aslan, Emrah; Almalki, Mishari Metab; Mossa, Mahmoud A.This study proposes an innovative stacking ensemble learning framework for classifying faults in induction motors utilized in Electric Vehicles (EVs). Employing a comprehensive dataset comprising motor data, such as speed, torque, current, and voltage, the analysis encompasses six distinct conditions: normal operating mode, over-voltage fault, under-voltage fault, overloading fault, phase-to-phase fault, and phase-to-ground fault. The proposed model integrates Gradient Boosting (GB), K-Nearest Neighbors (KNN), Gradient Boosting (XGBoost), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF) algorithms in a synergistic manner. The findings reveal that the RF-GB-DT-XGBoost combination achieves a remarkable accuracy of 98.53%, significantly surpassing other methods reported in the literature. Performance is evaluated through metrics including accuracy, precision, sensitivity, and F1-score, with results analyzed in comparison to practical applications and existing studies. Validated with real-world data, this study demonstrates that the proposed model offers a groundbreaking solution for predictive maintenance systems in the EV industry, exhibiting high generalization capacity despite complex operating conditions. This approach holds transformative potential for both academic research and industrial applications. The dataset used in this study was generated using a MATLAB 2018/Simulink-based Variable Frequency Drive (VFD) model that emulates real-world EV operating conditions rather than relying solely on laboratory data. This ensures that the developed model accurately reflects practical electric vehicle environments.Article Citation - WoS: 2Citation - Scopus: 2Comparison and Optimization of Machine Learning Methods for Fault Detection in District Heating and Cooling Systems(Polska Akad Nauk, Polish Acad Sci, Div Iv Technical Sciences Pas, 2025) Cinar, Mehmet; Aslan, Emrah; Ozupak, YildirimIn this study, the methods used for the detection of sub-station pollution failures in district heating and cooling (DHC) systems are analyzed. In the study, high, medium, and low-level pollution situations are considered and machine learning methods are applied for the detection of these failures. Random forest, decision tree, logistic regression, and CatBoost regression algorithms are compared within the scope of the analysis. The models are trained to perform fault detection at different pollution levels. To improve the model performance, hyper parameter optimization was performed with random search optimization, and the most appropriate values were selected. The results show that the CatBoost regression algorithm provides the highest accuracy and overall performance compared to other methods. The CatBoost model stood out with an accuracy of 0.9832 and a superior performance. These findings reveal that CatBoost-based approaches provide an effective solution in situations requiring high accuracy, such as contamination detection in DHC systems. The study makes an important contribution as a reliable fault detection solution in industrial applications.Article Citation - Scopus: 1Development of Malaria Diagnosis With Convolutional Neural Network Architectures: a Cnn-Based Software for Accurate Cell Image Analysis(Galileo Institute of Technology and Education of the Amazon (ITEGAM), 2025) Aslan, E.This study emphasizes that early diagnosis and treatment of malaria is critical in reducing health problems and mortality from the disease, especially in developing countries where the disease is prevalent. Malaria is a potentially fatal disease transmitted to humans by mosquitoes infected by a blood parasite called Plasmodium. The traditional method of diagnosis relies on experts examining red blood cells under a microscope and is inefficient as it is dependent on expert knowledge and experience. Nowadays, machine learning methods that provide high accuracy are increasingly used in disease detection. In this paper, a Convolutional Neural Network (CNN) architecture is proposed to distinguish between parasitized and non-parasitized cells. In addition, the performance of the proposed CNN architecture is compared to pre-trained CNN models such as VGG-19 and EfficientNetB3. The studies were carried out using the Malaria Dataset supplied by the National Institute of Health (NIH), and our proposed architecture was shown to function with 99.12% accuracy. The results of the study reveal that it is effective in improving the accuracy of cell images containing Plasmodium. In addition, a software that predicts whether cell images are noisy or not has been developed. © 2025 by authors and Galileo Institute of Technology and Education of the Amazon (ITEGAM).Article Citation - WoS: 6Citation - Scopus: 9Classification of Maize Leaf Diseases With Deep Learning: Performance Evaluation of the Proposed Model and Use of Explicable Artificial Intelligence(Elsevier, 2025) Alpsalaz, Feyyaz; Ozupak, Yildirim; Aslan, Emrah; Uzel, HasanMaize leaf diseases pose significant threats to global agricultural productivity, yet traditional diagnostic methods are slow, subjective, and resource-intensive. This study proposes a lightweight and interpretable convolutional neural network (CNN) model for accurate and efficient classification of maize leaf diseases. Using the 'Corn or Maize Leaf Disease Dataset', the model classifies four disease categories Healthy, Gray Leaf Spot, Common Rust, and Northern Leaf Blight with 94.97 % accuracy and a micro-average AUC of 0.99. With only 1.22 million parameters, the model supports real-time inference on mobile devices, making it ideal for field applications. Data augmentation and transfer learning techniques were applied to ensure robust generalization. To enhance transparency and user trust, Explainable Artificial Intelligence (XAI) methods, including LIME and SHAP, were employed to identify disease-relevant features such as lesions and pustules, with SHAP achieving an IoU of 0.82. The proposed model outperformed benchmark models like ResNet50, MobileNetV2, and EfficientNetB0 in both accuracy and computational efficiency. Robustness tests under simulated environmental challenges confirmed its adaptability, with only a 2.82 % performance drop under extreme conditions. Comparative analyses validated its statistical significance and practical superiority. This model represents a reliable, fast, and explainable solution for precision agriculture, especially in resource-constrained environments. Future enhancements will include multi-angle imaging, multimodal inputs, and extended datasets to improve adaptability and scalability in realworld conditions.Article Citation - WoS: 2Citation - Scopus: 4Hybrid Deep Learning Model for Maize Leaf Disease Classification With Explainable AI(Taylor & Francis Ltd, 2025) Ozupak, Yildirim; Alpsalaz, Feyyaz; Aslan, Emrah; Uzel, HasanThis study presents a hybrid learning model that integrates MobileNetV2 and Vision Transformer (ViT) with a stacking model to classify maize leaf diseases, addressing the critical need for early detection to improve agricultural productivity and sustainability. Utilising the 'Corn or Maize Leaf Disease Dataset' from Kaggle, comprising 4,062 high-resolution images across five classes (Common Rust, Grey Leaf Spot, Healthy, Northern Leaf Blight, Not Maize Leaf), the model achieves an impressive accuracy of 96.73%. Transfer learning from ImageNet, coupled with data augmentation (rotation, flipping, scaling, brightness adjustment), enhances generalisation, while a 20% dropout rate mitigates overfitting. The key advantage of the hybrid model lies in its ability to combine the strengths of MobileNetV2's localised feature extraction and ViT's global context understanding, enhanced by the stacking model's ability to reduce the weaknesses of either model. Explainable AI techniques, including SHapley Additive exPlanations (SHAP), Local Interpretable Model-agnostic Explanations (LIME), and Grad-CAM, provide transparent insights into model decisions, fostering trust among agricultural stakeholders. Comparative analysis demonstrates the model's superiority over prior works, with F1-scores ranging from 0.9276 to 1.0000. Despite minor misclassifications due to visual similarities, the model offers a robust, interpretable solution for precision agriculture.Article Citation - Scopus: 1Alzheimer’s Classification with a MaxViT-Based Deep Learning Model Using Magnetic Resonance Imaging(Interdisciplinary Publishing Academia, 2025) Demirtaş Alpsalaz, S.; Aslan, E.; Özüpak, Y.; Alpsalaz, F.; Uzel, H.Alzheimer’s disease (AD), a progressive neurodegenerative disorder, poses significant challenges for early diagnosis due to subtle symptom onset and overlap with normal aging. This study aims to develop an effective deep learning model for classifying four AD stages (Non-Demented, Very Mild Demented, Mild Demented, Moderate Demented) using brain MRI scans. We propose a Multi-Axis Vision Transformer (MaxViT)-based framework, leveraging transfer learning and robust data augmentation on the Kaggle Alzheimer’s MRI Dataset to address class imbalance and enhance generalization. The model employs MaxViT’s multi-axis attention mechanisms to capture both local and global patterns in MRI images. Our approach achieved a classification accuracy of 99.60%, with precision of 99.0%, recall of 98.1%, and F1-score of 98.51%. These results highlight MaxViT’s superior ability to differentiate AD stages, particularly in distinguishing challenging early stages. The proposed model offers a reliable tool for early AD diagnosis, laying a strong foundation for future clinical applications and interdisciplinary research in neurodegenerative disease detection. Future work should explore larger, more diverse datasets and additional biomarkers to further validate and enhance model performance. © 2025 Elsevier B.V., All rights reserved.Article Citation - WoS: 10Citation - Scopus: 12Air Quality Forecasting Using Machine Learning: Comparative Analysis and Ensemble Strategies for Enhanced Prediction(Springer Int Publ Ag, 2025) Ozupak, Yildirim; Alpsalaz, Feyyaz; Aslan, EmrahAir pollution poses a critical challenge to environmental sustainability, public health, and urban planning. Accurate air quality prediction is essential for devising effective management strategies and early warning systems. This study utilized a dataset comprising hourly measurements of pollutants such as PM2.5, NOx, CO, and benzene, sourced from five metal oxide sensors and a certified analyzer in a polluted urban area, totaling 9,357 records collected over one year (March 2004-February 2005) from the Kaggle Air Quality Data Set. A comprehensive comparison of ten machine learning regression models XGBoost, LightGBM, Random Forest, Gradient Boosting, CatBoost, Support Vector Regression (SVR) with Bayesian Optimization, Decision Tree, K-Nearest Neighbors (KNN), Elastic Net, and Bayesian Ridge was conducted. Model performance was enhanced through Bayesian optimization and randomized cross-validation, with stacking employed to leverage the strengths of base models. Experimental results showed that hyperparameter optimization and ensemble strategies significantly improved accuracy, with the SVR model optimized via Bayesian optimization achieving the highest performance: an R2 score of 99.94%, MAE of 0.0120, and MSE of 0.0005. These findings underscore the methodology's efficacy in precisely capturing the spatial and temporal dynamics of air pollution.Article Citation - WoS: 1Citation - Scopus: 1Boiler Efficiency and Performance Optimization in District Heating and Cooling Systems With Machine Learning Models(Taylor & Francis Ltd, 2025) Aslan, Emrah; Oezuepak, Yildirim; Alpsalaz, FeyyazThis study focuses on the detection and analysis of boiler efficiency degradation in District Heating and Cooling (DHC) substations. The research presents an innovative approach to optimize boiler efficiency under different scenarios. Although DHC systems provide both heating and cooling services, this study focuses specifically on heating substations. In this context, various machine learning algorithms have been applied to effectively detect boiler efficiency degradation, and hyper-parameter adjustments have been performed using Bayesian optimization to improve the performance of the models. As a result of the analyses, the Gradient Boosting Regressor model showed significantly higher performance compared to other machine learning algorithms. The model successfully predicted the decline in boiler efficiency with an accuracy of 97.8%, and the Matthews Correlation Coefficient (MCC) value was recorded as 0.952. These results show that Gradient Boosting Regressor based approaches provide an effective solution for fault detection and diagnosis in district heating systems. In conclusion, this study provides both theoretical and practical contributions to the optimization of boiler efficiency, fault detection and diagnosis in DHC systems. The solutions offered by the study have the potential to increase the reliability and efficiency of the systems.Article Hybrid Deep Learning with Attention Fusion for Enhanced Colon Cancer Detection(Nature Portfolio, 2025) Alpsalaz, Suheyla Demirtas; Aslan, Emrah; Ozupak, Yildirim; Alpsalaz, Feyyaz; Uzel, Hasan; Bereznychenko, ViktoriaThis study introduces a hybrid deep learning model integrating EfficientNet-B3 and Vision Transformer with an Attention Fusion mechanism for automated colon cancer detection using the Kvasir endoscopic dataset. The model leverages EfficientNet-B3's strength in capturing fine-grained local textures and Vision Transformer's ability to model global contextual relationships. A multi-head attention-based fusion block harmonizes these features, achieving comprehensive representations and enhanced classification stability. Model optimization was guided by the Matthews Correlation Coefficient (MCC), alongside evaluations of accuracy, F1-score, and Brier Score. Experimental results demonstrate a 96.2% accuracy and an MCC of 0.961, surpassing standalone baselines and existing benchmark architectures. Cross-validation confirmed robust generalization, while Grad-CAM analyses improved interpretability by visualizing salient histopathological regions influencing predictions. Despite slight overfitting tendencies, the model maintained strong performance across all eight image classes. These findings highlight the model's ability to address limitations of single-architecture approaches by combining local and global feature extraction, offering rapid, objective, and reliable diagnostic support. The proposed framework shows significant promise for integration into computer-aided colonoscopy systems, paving the way for enhanced clinical diagnostics and reduced pathologist workload through AI-driven precision medicine.
