Elif Aktepe
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Elif Aktepe has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Uçman Ergün
Engineering Faculty, Biomedical Engineering Department, Afyon Kocatepe University, Afyonkarahisar 03200, Turkey
3 shared publications - Yavuz Bahadır Koca
Electrical Engineering Department, Faculty of Engineering, Afyon Kocatepe University, Afyonkarahisar, Turkey
3 shared publications - Murat Tekbaş
Afyon Vocational School, Management and Organization Department, Afyon Kocatepe University, Afyonkarahisar, Turkey
2 shared publications - Şükriye Aktepe
Afyon Kocatepe Universitesi
1 shared publication
Latest Publications
- Open Access
- 13 October 2026
- Published by Özgür Yayınları
- p. 163-192
Digital Transformation in Healthcare: The Impacts of Artificial Intelligence on Healthcare Services and Health Economics
Digital transformation in healthcare, driven by the widespread adoption of electronic health records, the growth of large-scale health data, the expansion of telehealth applications, and advances in data-processing capacity, is strengthening the role of artificial intelligence within healthcare systems. Technologies such as machine learning, deep learning, natural language processing, and generative artificial intelligence are increasingly used in diagnosis and screening, clinical decision support, personalized care, patient monitoring, workflow optimization, and the improvement of access to healthcare services. This chapter provides a comprehensive conceptual and narrative review of the effects of artificial intelligence on healthcare services and health economics. It examines the impact of AI on diagnostic accuracy, early detection, clinical decision-making, healthcare professionals’ workload, service capacity, workforce productivity, access, and equity. In addition, the economic value of AI applications is evaluated in terms of cost-effectiveness, cost-benefit, resource allocation, reimbursement mechanisms, health technology assessment, and implementation costs. Existing evidence indicates that AI can accelerate diagnostic processes, reduce unnecessary procedures and tests, improve the utilization of healthcare professionals and high-cost medical equipment, and generate cost-effective outcomes in specific applications. However, the economic value of AI is not uniform across healthcare systems or areas of application. Patient volume, workforce costs, existing infrastructure, licensing and integration expenses, maintenance and updating costs, reimbursement structures, and institutional characteristics of healthcare systems substantially influence economic outcomes. Ethical and regulatory issues, including data security, algorithmic bias, explainability, clinical accountability, human oversight, and equity in access to healthcare, are also critical to the sustainable adoption of AI technologies. Overall, the value of artificial intelligence in healthcare should not be assessed solely on the basis of technical performance or short-term cost savings. A comprehensive assessment should consider clinical outcomes, economic value, implementation conditions, human–AI collaboration, ethical responsibilities, and long-term sustainability together.
- 13 October 2026
- Published by Özgür Yayınları
- p. 163-192
Open AccessDigital Transformation in Healthcare: The Impacts of Artificial Intelligence on Healthcare Services and Health Economics
Digital transformation in healthcare, driven by the widespread adoption of electronic health records, the growth of large-scale health data, the expansion of telehealth applications, and advances in data-processing capacity, is strengthening the role of artificial intelligence within healthcare systems. Technologies such as machine learning, deep learning, natural language processing, and generative artificial intelligence are increasingly used in diagnosis and screening, clinical decision support, personalized care, patient monitoring, workflow optimization, and the improvement of access to healthcare services. This chapter provides a comprehensive conceptual and narrative review of the effects of artificial intelligence on healthcare services and health economics. It examines the impact of AI on diagnostic accuracy, early detection, clinical decision-making, healthcare professionals’ workload, service capacity, workforce productivity, access, and equity. In addition, the economic value of AI applications is evaluated in terms of cost-effectiveness, cost-benefit, resource allocation, reimbursement mechanisms, health technology assessment, and implementation costs. Existing evidence indicates that AI can accelerate diagnostic processes, reduce unnecessary procedures and tests, improve the utilization of healthcare professionals and high-cost medical equipment, and generate cost-effective outcomes in specific applications. However, the economic value of AI is not uniform across healthcare systems or areas of application. Patient volume, workforce costs, existing infrastructure, licensing and integration expenses, maintenance and updating costs, reimbursement structures, and institutional characteristics of healthcare systems substantially influence economic outcomes. Ethical and regulatory issues, including data security, algorithmic bias, explainability, clinical accountability, human oversight, and equity in access to healthcare, are also critical to the sustainable adoption of AI technologies. Overall, the value of artificial intelligence in healthcare should not be assessed solely on the basis of technical performance or short-term cost savings. A comprehensive assessment should consider clinical outcomes, economic value, implementation conditions, human–AI collaboration, ethical responsibilities, and long-term sustainability together.
- Open Access
- 22 July 2026
- Published by Springer Nature in BMC Public Health
- Vol. 26 (1)
Predicting health, economy, and environmental indicators: a comparative machine learning approach
Background: This study aims to reveal the interaction between environmental sustainability and health policies by examining the relationships between economic growth, carbon emissions, and healthcare expenditures within the framework of machine learning methods. The primary motivation of the research is to evaluate the relationships among economic growth, health indicators, and environmental sustainability using data-driven methods. Methods: The analysis is based on a large-scale panel dataset covering 217 countries for the period 1960–2024, obtained from the World Bank World Development Indicators database. Advanced machine learning algorithms, including Extra Trees, Random Forest, Extreme Gradient Boosting, Histogram-based Gradient Boosting, and Light Gradient Boosting Machine, were implemented, and their prediction performances were compared. To improve prediction accuracy, a hybrid ensemble model was proposed by combining the highest-performing algorithms. For model interpretability, the relative importance levels of variables were analyzed using the Explainable Artificial Intelligence approach via Shapley Additive Explanations. Results: The findings indicate that the proposed hybrid ensemble model provides higher prediction accuracy compared to individual algorithms. Results from the Shapley Additive Explanations analysis show that health expenditures, economic growth, and innovation-related indicators exhibited high predictive importance within the developed machine learning models. The findings highlight the predictive associations between environmental indicators, healthcare spending, and sustainable development outcomes. Conclusions: This study analyzes the complex relationships between economic growth, environmental sustainability, and health policies through a data-driven perspective. The findings provide additional evidence regarding the predictive associations among economic, environmental, and health-related indicators. The proposed framework demonstrates the usefulness of machine learning approaches for analyzing complex nonlinear patterns in large-scale country-year datasets.
- 22 July 2026
- Published by Springer Nature in BMC Public Health
- Vol. 26 (1)
Open AccessPredicting health, economy, and environmental indicators: a comparative machine learning approach
Background: This study aims to reveal the interaction between environmental sustainability and health policies by examining the relationships between economic growth, carbon emissions, and healthcare expenditures within the framework of machine learning methods. The primary motivation of the research is to evaluate the relationships among economic growth, health indicators, and environmental sustainability using data-driven methods. Methods: The analysis is based on a large-scale panel dataset covering 217 countries for the period 1960–2024, obtained from the World Bank World Development Indicators database. Advanced machine learning algorithms, including Extra Trees, Random Forest, Extreme Gradient Boosting, Histogram-based Gradient Boosting, and Light Gradient Boosting Machine, were implemented, and their prediction performances were compared. To improve prediction accuracy, a hybrid ensemble model was proposed by combining the highest-performing algorithms. For model interpretability, the relative importance levels of variables were analyzed using the Explainable Artificial Intelligence approach via Shapley Additive Explanations. Results: The findings indicate that the proposed hybrid ensemble model provides higher prediction accuracy compared to individual algorithms. Results from the Shapley Additive Explanations analysis show that health expenditures, economic growth, and innovation-related indicators exhibited high predictive importance within the developed machine learning models. The findings highlight the predictive associations between environmental indicators, healthcare spending, and sustainable development outcomes. Conclusions: This study analyzes the complex relationships between economic growth, environmental sustainability, and health policies through a data-driven perspective. The findings provide additional evidence regarding the predictive associations among economic, environmental, and health-related indicators. The proposed framework demonstrates the usefulness of machine learning approaches for analyzing complex nonlinear patterns in large-scale country-year datasets.
- Open Access
- 18 July 2026
- Published by MDPI AG in Micromachines
- Vol. 17 (7) , 859
HMQ-ES-Stack-GBR: A Hybrid Ensemble Learning Model for Mechanical and Physical Quality Prediction in FDM 3D Printing
In Fusion Deposition Modeling-based manufacturing, process parameters affect the mechanical and physical properties of the print. Considering these properties, accurately predicting print quality is essential. This is where machine learning (ML) models for three-dimensional (3D) print quality prediction come to the forefront. In this study, a dataset was prepared under strict operational measurement standards—utilizing the Interquartile Range (IQR) method for data sanitization—encompassing 10 material types, 2 printer types, and 4 printing parameters. Five hundred different sample combinations were prepared and printed in sets of three according to ISO 527-2 Type 4 standard dimensions. Tensile, hardness, and surface roughness tests were applied to the prepared samples. Using this validated dataset, a Hybrid Multi-Material Quality–Ensemble System–Stacking–Gradient Boosting Regressor (HMQ-ES-Stack-GBR) architecture is proposed as a diagnostic framework for multi-output quality prediction. Particularly in terms of quality outputs such as tensile strength, hardness, and surface roughness, while also providing a quantitative analysis of the effect of material type on print quality. Furthermore, a multi-objective optimization pipeline integrating three distinct meta-heuristic algorithms—Non-dominated Sorting Genetic Algorithm II (NSGA-II), Particle Swarm Optimization (PSO), and Grey Wolf Optimizer (GWO)—was coupled with the framework to systematically derive material-specific optimal processing parameter configurations. Furthermore, the study shows that open-system printers exhibit higher prediction errors than closed-system printers. Reflecting system-induced variability rather than full hardware independence. Although the study is limited to internal validation within the current experimental design and includes material imbalance across filament groups, the findings suggest that the proposed framework provides a promising diagnostic decision-support tool for pre-print quality estimation within the studied dataset. By accurately reflecting rather than physically overcoming manufacturing variability, it supports decision-making processes through pre-print quality estimation, thereby enabling proactive interventions that reduce raw material, time, and energy losses.
- 18 July 2026
- Published by MDPI AG in Micromachines
- Vol. 17 (7) , 859
Open AccessHMQ-ES-Stack-GBR: A Hybrid Ensemble Learning Model for Mechanical and Physical Quality Prediction in FDM 3D Printing
In Fusion Deposition Modeling-based manufacturing, process parameters affect the mechanical and physical properties of the print. Considering these properties, accurately predicting print quality is essential. This is where machine learning (ML) models for three-dimensional (3D) print quality prediction come to the forefront. In this study, a dataset was prepared under strict operational measurement standards—utilizing the Interquartile Range (IQR) method for data sanitization—encompassing 10 material types, 2 printer types, and 4 printing parameters. Five hundred different sample combinations were prepared and printed in sets of three according to ISO 527-2 Type 4 standard dimensions. Tensile, hardness, and surface roughness tests were applied to the prepared samples. Using this validated dataset, a Hybrid Multi-Material Quality–Ensemble System–Stacking–Gradient Boosting Regressor (HMQ-ES-Stack-GBR) architecture is proposed as a diagnostic framework for multi-output quality prediction. Particularly in terms of quality outputs such as tensile strength, hardness, and surface roughness, while also providing a quantitative analysis of the effect of material type on print quality. Furthermore, a multi-objective optimization pipeline integrating three distinct meta-heuristic algorithms—Non-dominated Sorting Genetic Algorithm II (NSGA-II), Particle Swarm Optimization (PSO), and Grey Wolf Optimizer (GWO)—was coupled with the framework to systematically derive material-specific optimal processing parameter configurations. Furthermore, the study shows that open-system printers exhibit higher prediction errors than closed-system printers. Reflecting system-induced variability rather than full hardware independence. Although the study is limited to internal validation within the current experimental design and includes material imbalance across filament groups, the findings suggest that the proposed framework provides a promising diagnostic decision-support tool for pre-print quality estimation within the studied dataset. By accurately reflecting rather than physically overcoming manufacturing variability, it supports decision-making processes through pre-print quality estimation, thereby enabling proactive interventions that reduce raw material, time, and energy losses.
- 16 June 2026
- Published by Journal of Materials and Mechatronics: A in Journal of Materials and Mechatronics: A
- Vol. 7 (1) , 109-121
Deep Learning Application for Image-Based Defect Detection in 3D Printing Processes
Fused Deposition Modeling (FDM) is a widely preferred additive manufacturing technology due to its design flexibility and cost advantages. However, structural and visual defects such as stringing, layer shifting, warping, or off-platform issues frequently occur in FDM-based three-dimensional (3D) printing processes. This results in losses of raw materials, energy, and time. This study aims to classify common defect types occurring in FDM-based 3D printing using image-based deep learning algorithms. In this study, a transfer learning approach is adopted using the EfficientNetB0 and MobileNetV2 architectures, both pre-trained on ImageNet, with a dataset of 1912 images. The model’s generalization capability is enhanced through data preprocessing and enhancement techniques. The model can successfully classify defect types using print images without additional sensors or hardware components. The results show that the EfficientNetB0 model has an overall accuracy of 87.7%, while MobileNetV2 achieves 97%. The MobileNetV2 architecture demonstrates strong performance, particularly in error classes such as Layer Shifting and Stringing, with high F1 scores. The proposed architectures have the potential to reduce losses in FDM-based 3D printing processes by providing a low-cost and accessible visual quality control system.
- 16 June 2026
- Published by Journal of Materials and Mechatronics: A in Journal of Materials and Mechatronics: A
- Vol. 7 (1) , 109-121
Deep Learning Application for Image-Based Defect Detection in 3D Printing Processes
Fused Deposition Modeling (FDM) is a widely preferred additive manufacturing technology due to its design flexibility and cost advantages. However, structural and visual defects such as stringing, layer shifting, warping, or off-platform issues frequently occur in FDM-based three-dimensional (3D) printing processes. This results in losses of raw materials, energy, and time. This study aims to classify common defect types occurring in FDM-based 3D printing using image-based deep learning algorithms. In this study, a transfer learning approach is adopted using the EfficientNetB0 and MobileNetV2 architectures, both pre-trained on ImageNet, with a dataset of 1912 images. The model’s generalization capability is enhanced through data preprocessing and enhancement techniques. The model can successfully classify defect types using print images without additional sensors or hardware components. The results show that the EfficientNetB0 model has an overall accuracy of 87.7%, while MobileNetV2 achieves 97%. The MobileNetV2 architecture demonstrates strong performance, particularly in error classes such as Layer Shifting and Stringing, with high F1 scores. The proposed architectures have the potential to reduce losses in FDM-based 3D printing processes by providing a low-cost and accessible visual quality control system.
- 14 October 2025
- Published by Springer Nature
- p. 277-286
Optimization of 3D Printing Parameters Using Machine Learning Techniques
- 14 October 2025
- Published by Springer Nature
- p. 277-286
Optimization of 3D Printing Parameters Using Machine Learning Techniques
- Open Access
- 12 September 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (18) , 10001
Machine Learning Approaches for FDM-Based 3D Printing: A Literature Review
Three-dimensional (3D) printing has become a widely used manufacturing technology, but predicting the mechanical and physical properties of printed parts remains a critical challenge. In particular, tensile strength, hardness, and surface roughness are essential for assessing product quality and reliability. Addressing this gap requires advanced computational approaches. Machine Learning (ML) algorithms have the potential to enhance automation and provide accurate predictions of product quality in 3D printing. The primary objective of this study is to review, classify, and analyze recent research on the application of ML algorithms for predicting mechanical (tensile strength, hardness) and physical (surface roughness) properties in 3D-printed materials. This review aims to identify current research trends, highlight existing gaps in the literature, and propose potential directions for future investigations in this rapidly evolving interdisciplinary field. For this purpose, a literature review was conducted in the Web of Science database using keywords such as “3D printing”, “machine learning”, “tensile strength”, “hardness”, and “roughness. The review systematically examines the material types, ML algorithms, printing parameters, and testing methods reported in the literature, and the frequency of these parameters is visualized through graphs to illustrate research trends. The findings reveal that ML algorithms achieve high accuracy in predicting tensile strength, hardness, and surface roughness, providing valuable insights for process optimization. However, current research still shows limited evidence for direct improvements in efficiency and error reduction, indicating important directions for future work.
- 12 September 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (18) , 10001
Open AccessMachine Learning Approaches for FDM-Based 3D Printing: A Literature Review
Three-dimensional (3D) printing has become a widely used manufacturing technology, but predicting the mechanical and physical properties of printed parts remains a critical challenge. In particular, tensile strength, hardness, and surface roughness are essential for assessing product quality and reliability. Addressing this gap requires advanced computational approaches. Machine Learning (ML) algorithms have the potential to enhance automation and provide accurate predictions of product quality in 3D printing. The primary objective of this study is to review, classify, and analyze recent research on the application of ML algorithms for predicting mechanical (tensile strength, hardness) and physical (surface roughness) properties in 3D-printed materials. This review aims to identify current research trends, highlight existing gaps in the literature, and propose potential directions for future investigations in this rapidly evolving interdisciplinary field. For this purpose, a literature review was conducted in the Web of Science database using keywords such as “3D printing”, “machine learning”, “tensile strength”, “hardness”, and “roughness. The review systematically examines the material types, ML algorithms, printing parameters, and testing methods reported in the literature, and the frequency of these parameters is visualized through graphs to illustrate research trends. The findings reveal that ML algorithms achieve high accuracy in predicting tensile strength, hardness, and surface roughness, providing valuable insights for process optimization. However, current research still shows limited evidence for direct improvements in efficiency and error reduction, indicating important directions for future work.
- Open Access
- 26 September 2024
- Published by Bingol Universitesi in Türk Doğa ve Fen Dergisi
- Vol. 13 (3) , 109-116
Evaluation of Missing Data Imputation Methods and PCA Techniques for Machine Learning Models in Breast Cancer Diagnosis Using WBCD
Cancer is one of the leading causes of human mortality and breast cancer deaths are particularly common among women. Early diagnosis of breast cancer is considered a key way to reduce these deaths. The use of expert systems, artificial intelligence and machine learning techniques in the medical field aims to assist doctors in early disease detection. One of the main objectives of these technologies is to diagnose life-threatening diseases such as breast cancer earlier and more accurately. In this study, analyses conducted on the Wisconsin Breast Cancer Dataset (WBCD) evaluated the effects of different missing data imputation methods and PCA-based data reduction technique on model performance using supervised machine learning methods. In the first stage of the study, the detection and management of missing values in the dataset were addressed. It was found that imputing missing values with median performed better compared to other methods. Subsequently, the dataset was reduced in size using the PCA method and the performance of algorithms was measured by experimenting with different numbers of components. The results indicate that effectively addressing the missing data problem and using PCA-based data reduction techniques significantly improve model performance. The best performance was achieved by imputing missing data with median values and reducing data dimensionality with PCA. This study emphasizes the importance of combining machine learning approaches for breast cancer diagnosis with missing data management strategies. Additionally, the effects of different missing data imputation methods and PCA on model performance have been thoroughly examined.
- 26 September 2024
- Published by Bingol Universitesi in Türk Doğa ve Fen Dergisi
- Vol. 13 (3) , 109-116
Open AccessEvaluation of Missing Data Imputation Methods and PCA Techniques for Machine Learning Models in Breast Cancer Diagnosis Using WBCD
Cancer is one of the leading causes of human mortality and breast cancer deaths are particularly common among women. Early diagnosis of breast cancer is considered a key way to reduce these deaths. The use of expert systems, artificial intelligence and machine learning techniques in the medical field aims to assist doctors in early disease detection. One of the main objectives of these technologies is to diagnose life-threatening diseases such as breast cancer earlier and more accurately. In this study, analyses conducted on the Wisconsin Breast Cancer Dataset (WBCD) evaluated the effects of different missing data imputation methods and PCA-based data reduction technique on model performance using supervised machine learning methods. In the first stage of the study, the detection and management of missing values in the dataset were addressed. It was found that imputing missing values with median performed better compared to other methods. Subsequently, the dataset was reduced in size using the PCA method and the performance of algorithms was measured by experimenting with different numbers of components. The results indicate that effectively addressing the missing data problem and using PCA-based data reduction techniques significantly improve model performance. The best performance was achieved by imputing missing data with median values and reducing data dimensionality with PCA. This study emphasizes the importance of combining machine learning approaches for breast cancer diagnosis with missing data management strategies. Additionally, the effects of different missing data imputation methods and PCA on model performance have been thoroughly examined.
- Open Access
- 30 April 2024
- Published by International Journal of 3D Printing Technologies and Digital Industry in International Journal of 3D Printing Technologies and Digital Industry
- Vol. 8 (1) , 114-123
PLA ve GERİ DÖNÜŞTÜRÜLMÜŞ PET FİLAMENTLERİNİN 3D FDM BASKIDA BOYUTSAL DOĞRULUK ve GERİÇEKİLME PERFORMANSININ KARŞILAŞTIRILMASI
Bu çalışmada PLA filament ve PET su şişesinin geri dönüştürülmesiyle üretilen PET filamentin boyutsal doğruluğunu ve geri çekilme performansını değerlendirmek için 3 boyutlu FDM (Fused Deposition Modeling) yazıcıda test objeleri baskısı alınmıştır. Çalışma, geri dönüştürülmüş PET filamentin kullanılabilirliğini ve performansını belirlemeyi hedeflemektedir. Araştırmada farklı baskı hızları ile PLA filament ve geri dönüştürülmüş PET filament kullanılarak baskı işlemleri gerçekleştirilmiştir. 3 boyutlu yazıcı ile farklı hızlarda (30-40-50 mm/s) üretilen test objelerinin boyutsal doğruluğu için ölçümler yapılmıştır. Bu ölçümler, filamentin yapısal tutarlılığını ve basılan nesnelerin boyutsal hassasiyetini değerlendirmek için yapılmaktadır. Geri çekilme performansı için de aynı baskı hızları kullanılarak hem PLA filament hem de geri dönüştürülmüş PET filament ile baskılar alınmıştır. Araştırmanın sonuçları, geri dönüştürülmüş PET filamentin 3 boyutlu FDM tipi yazıcılarda nasıl performans gösterdiğini ortaya koyarak, sürdürülebilir üretim sürecine katkıda bulunabilir.
- 30 April 2024
- Published by International Journal of 3D Printing Technologies and Digital Industry in International Journal of 3D Printing Technologies and Digital Industry
- Vol. 8 (1) , 114-123
Open AccessPLA ve GERİ DÖNÜŞTÜRÜLMÜŞ PET FİLAMENTLERİNİN 3D FDM BASKIDA BOYUTSAL DOĞRULUK ve GERİÇEKİLME PERFORMANSININ KARŞILAŞTIRILMASI
Bu çalışmada PLA filament ve PET su şişesinin geri dönüştürülmesiyle üretilen PET filamentin boyutsal doğruluğunu ve geri çekilme performansını değerlendirmek için 3 boyutlu FDM (Fused Deposition Modeling) yazıcıda test objeleri baskısı alınmıştır. Çalışma, geri dönüştürülmüş PET filamentin kullanılabilirliğini ve performansını belirlemeyi hedeflemektedir. Araştırmada farklı baskı hızları ile PLA filament ve geri dönüştürülmüş PET filament kullanılarak baskı işlemleri gerçekleştirilmiştir. 3 boyutlu yazıcı ile farklı hızlarda (30-40-50 mm/s) üretilen test objelerinin boyutsal doğruluğu için ölçümler yapılmıştır. Bu ölçümler, filamentin yapısal tutarlılığını ve basılan nesnelerin boyutsal hassasiyetini değerlendirmek için yapılmaktadır. Geri çekilme performansı için de aynı baskı hızları kullanılarak hem PLA filament hem de geri dönüştürülmüş PET filament ile baskılar alınmıştır. Araştırmanın sonuçları, geri dönüştürülmüş PET filamentin 3 boyutlu FDM tipi yazıcılarda nasıl performans gösterdiğini ortaya koyarak, sürdürülebilir üretim sürecine katkıda bulunabilir.
- Open Access
- 1 January 2024
- Published by Turkish Journal of Engineering in Turkish Journal of Engineering
- Vol. 8 (3) , 447-456
Effect of dimension reduction with PCA and machine learning algorithms on diabetes diagnosis performance
Diabetes, a long-term metabolic disorder, causes persistently high blood sugar and presents a significant global health challenge. Early diagnosis is of vital importance in mitigating the effects of diabetes. This study aims to investigate diabetes diagnosis and risk prediction using a comprehensive diabetes dataset created in 2023. The dataset contains clinical and anthropometric data of patients. Data simplification was successfully applied to clean unnecessary information and reduce data dimensionality. Additionally, methods like Principal Component Analysis were applied to decrease the number of variables in the dataset. These analyses rendered the dataset more manageable and improved its performance. In this study, a dataset encompassing health data of a total of 100,000 individuals was utilized. This dataset consists of 8 input features and 1 output feature. The primary objective is to determine the algorithm that exhibits the best performance for diabetes diagnosis. There was no missing data during the data preprocessing stage, and the necessary transformations were carried out successfully. Nine different machine learning algorithms were applied to the dataset in this study. Each algorithm employed various modelling approaches to evaluate its performance in diagnosing diabetes. The results demonstrate that machine learning models are successful in predicting the presence of diabetes and the risk of developing it in healthy individuals. Particularly, the random forest model provided superior results across all performance metrics. This study provides significant findings that can shed light on future research in diabetes diagnosis and risk prediction. Dimensionality reduction techniques have proven to be valuable in data analysis and have highlighted the potential to facilitate diabetes diagnosis, thereby enhancing the quality of life for patients.
- 1 January 2024
- Published by Turkish Journal of Engineering in Turkish Journal of Engineering
- Vol. 8 (3) , 447-456
Open AccessEffect of dimension reduction with PCA and machine learning algorithms on diabetes diagnosis performance
Diabetes, a long-term metabolic disorder, causes persistently high blood sugar and presents a significant global health challenge. Early diagnosis is of vital importance in mitigating the effects of diabetes. This study aims to investigate diabetes diagnosis and risk prediction using a comprehensive diabetes dataset created in 2023. The dataset contains clinical and anthropometric data of patients. Data simplification was successfully applied to clean unnecessary information and reduce data dimensionality. Additionally, methods like Principal Component Analysis were applied to decrease the number of variables in the dataset. These analyses rendered the dataset more manageable and improved its performance. In this study, a dataset encompassing health data of a total of 100,000 individuals was utilized. This dataset consists of 8 input features and 1 output feature. The primary objective is to determine the algorithm that exhibits the best performance for diabetes diagnosis. There was no missing data during the data preprocessing stage, and the necessary transformations were carried out successfully. Nine different machine learning algorithms were applied to the dataset in this study. Each algorithm employed various modelling approaches to evaluate its performance in diagnosing diabetes. The results demonstrate that machine learning models are successful in predicting the presence of diabetes and the risk of developing it in healthy individuals. Particularly, the random forest model provided superior results across all performance metrics. This study provides significant findings that can shed light on future research in diabetes diagnosis and risk prediction. Dimensionality reduction techniques have proven to be valuable in data analysis and have highlighted the potential to facilitate diabetes diagnosis, thereby enhancing the quality of life for patients.