In-depth analysis of machine learning models and explainable artificial intelligence methods in diabetes diagnosis

Hakan Güler; Derya Avcı; Mustafa Ulaş; Tülay Omma

Makale detayları

In-depth analysis of machine learning models and explainable artificial intelligence methods in diabetes diagnosis

Yazarlar:

Hakan Güler | Derya Avcı | Mustafa Ulaş | Tülay Omma

Yayıncı:

Gazi University

Ana Başlıklar:

Yapay Zeka (Diğer)

Dergi:

Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi Vol.40 Issue 3, pp.1995-2012

Belge Tipi:

Makale

Yayın Tarihi:

8/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001345033

pISSN:

1300-1884

Anahtar kelimeler:

Özet:

 With the increase in large datasets in the healthcare sector in recent years, machine learning methods have started to play an important role in the discovery, analysis and prediction processes of diabetes datasets. This study focused on the early diagnosis of diabetes and examined the performance of various machine learning models and the effect of Explainable Artificial Intelligence XAI techniques. The performances of the models were compared using K-Nearest Neighbors KNN, Support Vector Machines SVM, Naive Bayes, Artificial Neural Networks ANN, Decision Trees, Random Forest and XGBoost algorithms on the diabetes dataset. The evaluation of the models was carried out with training and testing processes after the data cleaning and preprocessing stages. During this evaluation, success criteria such as accuracy, F1 score, sensitivity and originality were taken into consideration. Explainable artificial intelligence methods such as SHAP Shapley Additive Explanations and LIME Local Interpretable Model-Agnostic Explanations were applied to the outputs of the model that gave the best results, and the understandability of the decisions made by the model was ensured. These methods determine the most important features of the model and make the decision process more transparent. In order to understand the potential of these models in real-world applications, the analysis results were supported by expert opinions. KNN, SVM, Naive Bayes, ANN, Decision Trees, Random Forest and XGBoost algorithms were evaluated with 81.18, 75.38, 75.49, 74.83, 76.91, 91.68 and 98.91 accuracy rates, respectively.

 With the increase in large datasets in the healthcare sector in recent years, machine learning methods have started to play an important role in the discovery, analysis and prediction processes of diabetes datasets. This study focused on the early diagnosis of diabetes and examined the performance of various machine learning models and the effect of Explainable Artificial Intelligence XAI techniques. The performances of the models were compared using K-Nearest Neighbors KNN, Support Vector Machines SVM, Naive Bayes, Artificial Neural Networks ANN, Decision Trees, Random Forest and XGBoost algorithms on the diabetes dataset. The evaluation of the models was carried out with training and testing processes after the data cleaning and preprocessing stages. During this evaluation, success criteria such as accuracy, F1 score, sensitivity and originality were taken into consideration. Explainable artificial intelligence methods such as SHAP Shapley Additive Explanations and LIME Local Interpretable Model-Agnostic Explanations were applied to the outputs of the model that gave the best results, and the understandability of the decisions made by the model was ensured. These methods determine the most important features of the model and make the decision process more transparent. In order to understand the potential of these models in real-world applications, the analysis results were supported by expert opinions. KNN, SVM, Naive Bayes, ANN, Decision Trees, Random Forest and XGBoost algorithms were evaluated with 81.18, 75.38, 75.49, 74.83, 76.91, 91.68 and 98.91 accuracy rates, respectively.

 With the increase in large datasets in the healthcare sector in recent years, machine learning methods have started to play an important role in the discovery, analysis and prediction processes of diabetes datasets. This study focused on the early diagnosis of diabetes and examined the performance of various machine learning models and the effect of Explainable Artificial Intelligence XAI techniques. The performances of the models were compared using K-Nearest Neighbors KNN, Support Vector Machines SVM, Naive Bayes, Artificial Neural Networks ANN, Decision Trees, Random Forest and XGBoost algorithms on the diabetes dataset. The evaluation of the models was carried out with training and testing processes after the data cleaning and preprocessing stages. During this evaluation, success criteria such as accuracy, F1 score, sensitivity and originality were taken into consideration. Explainable artificial intelligence methods such as SHAP Shapley Additive Explanations and LIME Local Interpretable Model-Agnostic Explanations were applied to the outputs of the model that gave the best results, and the understandability of the decisions made by the model was ensured. These methods determine the most important features of the model and make the decision process more transparent. In order to understand the potential of these models in real-world applications, the analysis results were supported by expert opinions. KNN, SVM, Naive Bayes, ANN, Decision Trees, Random Forest and XGBoost algorithms were evaluated with 81.18, 75.38, 75.49, 74.83, 76.91, 91.68 and 98.91 accuracy rates, respectively.

 Sağlık sektöründe son yıllarda büyük veri kümelerinin artmasıyla birlikte, makine öğrenimi yöntemleri diyabet veri setlerinin keşfi, analizi ve tahmin süreçlerinde önemli bir rol oynamaya başlamıştır. Bu çalışmada, diyabetin erken teşhisine odaklanılmış ve çeşitli makine öğrenimi modellerinin performansı ile Açıklanabilir Yapay Zeka XAI tekniklerinin etkisi incelenmiştir. Diyabet veri seti üzerinde, K-En Yakın Komşular KNN, Destek Vektör Makineleri SVM, Naive Bayes, Yapay Sinir Ağları YSA, Karar Ağaçları, Rastgele Orman ve XGBoost algoritmaları kullanılarak modellerin performansları karşılaştırılmıştır. Modellerin değerlendirilmesi, veri temizleme ve ön işleme aşamalarının ardından eğitim ve test süreçleriyle yapılmıştır. Bu değerlendirme sırasında doğruluk, F1 skoru, hassasiyet ve özgünlük gibi başarı kriterleri göz önünde bulundurulmuştur. En iyi sonuç veren modelin çıktılarına SHAP Shapley Additive Explanations ve LIME Local Interpretable Model-Agnostic Explanations gibi açıklanabilir yapay zeka yöntemleri uygulanarak, modelin verdiği kararların anlaşılabilirliği sağlanmıştır. Bu yöntemler, modelin en önemli özelliklerini belirleyip karar sürecini daha şeffaf hale getirir. Gerçek dünya uygulamalarında bu modellerin potansiyelini anlamak adına uzman görüşleriyle analiz sonuçları desteklenmiştir. KNN, SVM, Naive Bayes, YSA, Karar Ağaçları, Rastgele Orman ve XGBoost algoritmaları sırasıyla 81.18, 75.38, 75.49, 74.83, 76.91, 91.68 ve 98.91 doğruluk oranlarıyla değerlendirilmiştir.

Makale detayları

In-depth analysis of machine learning models and explainable artificial intelligence methods in diabetes diagnosis

Yazarlar:

Hakan Güler | Derya Avcı | Mustafa Ulaş | Tülay Omma

Yayıncı:

Gazi University

Ana Başlıklar:

Yapay Zeka (Diğer)

Dergi:

Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi Vol.40 Issue 3, pp.1995-2012

Belge Tipi:

Makale

Yayın Tarihi:

8/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001345033

pISSN:

1300-1884

Diabetes

Explainable Artificial Intelligence

SHAP

XGBoostAçıklanabilir Yapay Zeka

Diyabet

SHAP

XGBoost

Sayı

Makale detayları

Yazarlar: Hakan Güler | Derya Avcı | Mustafa Ulaş | Tülay Omma

Yayıncı: Gazi University

Ana Başlıklar: Yapay Zeka (Diğer)

Dergi: Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi Vol.40 Issue 3, pp.1995-2012

Belge Tipi: Makale

Yayın Tarihi: 8/2025

Belge Dili: Türkçe

Document ID (Org No.): 2001345033

pISSN: 1300-1884

Sayı

Yazarlar:

Hakan Güler | Derya Avcı | Mustafa Ulaş | Tülay Omma

Yayıncı:

Gazi University

Ana Başlıklar:

Yapay Zeka (Diğer)

Dergi:

Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi Vol.40 Issue 3, pp.1995-2012

Belge Tipi:

Makale

Yayın Tarihi:

8/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001345033

pISSN:

1300-1884