Jul 2026· Journal of Innovative Engineering and Natural Science· Vol 6, pp. 399-413· 0 citations· 17 references
TL;DR
The results demonstrate that ensemble learning structures, which combine the strengths of multiple models, can provide more reliable decision support in critical areas such as healthcare and emphasise that ensemble learning structures, which combine the strengths of multiple models, can provide more reliable decision support in critical areas such as healthcare.
Abstract
The increasing global prevalence of diabetes mellitus poses severe public health challenges, underscoring the vital need for prompt and accurate diagnostic strategies. In this study, machine learning and deep learning-based models were individually evaluated for diabetes prediction, and the best-performing models were subsequently combined using a voting classifier structure. The dataset used was obtained from Kaggle, and preprocessing steps included the removal of missing values and outliers (resulting in the removal of 35,828 records, or 35.8% of the initial data). Statistical analysis showed that none of the continuous variables followed a normal distribution, and Mann-Whitney U tests revealed statistically significant differences between diabetic and non-diabetic groups. Five machine learning algorithms—Random Forest, XGBoost, Naive Bayes, Logistic Regression, and Support Vector Machine—and two deep learning models—Multilayer Perceptron and TabNet—were assessed based on accuracy, precision, recall, and F1-score. The top-performing models (RF, XGBoost, MLP, and TabNet) were integrated using a soft voting approach. The results demonstrate that the voting classifier exhibits highly balanced performance and achieves the highest scores among the compared models in terms of accuracy, precision, specificity, F1 score, and MCC metrics. However, it is notable that for the Recall metric-which is clinically critical for primary disease screening- the individual MLP (0.7000) and XGBoost (0.6964) models outperformed the ensemble structure. These findings emphasise that ensemble learning structures, which combine the strengths of multiple models, can provide more reliable decision support in critical areas such as healthcare. For future research, the generalizability of the model can be enhanced by using datasets with a broader range of features, and real-time adaptive systems supported by cloud computing technologies can be developed to improve the responsiveness and efficiency of diabetes prediction models.
Diabetes mellitus is a chronic metabolic disorder whose global prevalence continues to rise, creating an urgent need for scalable, low-cost tools for early risk identification. This study evaluates the effectiveness of five machine learning algorithms (Random Forest, XGBoost, Support Vector Machine [SVM], CatBoost, and TabNet) for predicting diabetes risk from routinely available clinical and lifestyle variables. Using the Pima Indians Diabetes dataset, a preprocessing pipeline was applied that included median imputation of physiologically implausible zero values, standardized (Z-score) feature scaling, Boruta-based feature selection, and class-imbalance handling through class-weight adjustment and the Synthetic Minority Oversampling Technique (SMOTE). Models were trained on an 80/20 train-test split and assessed using accuracy, precision, recall, F1-score, and the area under the receiver operating characteristic curve (ROC-AUC). Random Forest achieved the strongest overall performance (accuracy 0.753; F1-score 0.689; ROC-AUC 0.810), followed by CatBoost, SVM, and XGBoost, whereas TabNet performed worst with very low recall for the diabetic class. The best-performing model (Random Forest) was deployed in a lightweight Flask web application that returns a probability-based diabetes risk assessment, categorising each prediction as low, moderate, or high risk together with a tailored recommendation. The findings confirm that ensemble tree-based methods, particularly Random Forest, provide a reliable, interpretable, and deployable basis for diabetes risk screening, especially in resource-constrained settings. Key limitations include dataset homogeneity, residual class imbalance, and limited feature coverage.
T. Olayinka· FUDMA Journal of Sciences· 0 citations
The experimental results show that for the GNB model, the best performance was achieved using the combination of StandardScaler, SMOTE, and SelectKBest (k=5), reaching an accuracy of 94.53%, precision 98.36%, recall 90.91%, and f1-score 94.49%.
Joshua Roy Danna Lacanlale, Vitri Tundjungsari· International Journal of Sci...· 0 citations
Random Forest provided the best overall performance for three-class diabetes classification in this analytical sample, however, modest agreement, low prediabetes sensitivity, potential label leakage from fasting glucose, and the absence of external validation indicate that further evaluation is required before clinical application.
Ayeni Taiwo Michael, Odukoya Ayooluwa, Ilesanmi Opeyemi· Asian Research Journal of Ma...· 0 citations
Diabetes affects over 101 million people in India, with many more at risk due to routine and hereditary factors. Early diagnosis is crucial to prevent complications, which make accurate predictive tools essential in healthcare. This research uses Machine Learning (ML) algorithms to evaluate the likelihood of Type 2 Diabetes Mellitus (T2DM) using lifestyle and family history data. The trained models demonstrate strong predictive ability, allowing individuals to self-assess their risk and supporting healthcare professionals in early detection and intervention. This study presents a performance assessment of seven ML classifiers: Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), Logistic Regression (LR), Naïve Bayes (NB), k-Nearest Neighbor (k-NN), and Extreme Gradient Boosting (XGBoost). These classifiers were applied to the widely used PIMA Indian Diabetes dataset (PIDD), which contains 768 clinical records of adult women aged 21 and above, providing key medical information for diabetes analysis. Multiple evaluation measures were applied to assess model performance with results showing that SVM achieved the highest accuracy and AUC, while LR, RF, and XGBoost also performed competitively. Although k-NN attained the highest recall, it yielded a higher false positive rate. These findings highlight that no single model is perfect for every situation, and the choice of classifier should match clinical needs. This study serves as a reference for ML applications in diabetes prediction.
Rizwan Akhtar, Muhammad Kalamuddin Ahamad· ITEGAM- Journal of Engineeri...· 0 citations
It is recommended that healthcare systems adopt XGBoost-based predictive models in clinical decision support tools for early screening, while future studies should validate these models using real-world clinical data to enhance reliability and generalizability.
Idehen Emmanuel Imafidon, Chikere Obinna Munachiso, Dominic Evans Onyebuchi et al.· International Journal of Sci...· 0 citations
The study comes to the conclusion that headline accuracy is an unreliable guide in imbalanced medical prediction, that imbalance handling can change a model's practical usefulness, and that this benefit is strongly algorithm-dependent, meaning that the decision to resample should be based on the algorithm and the screening priorities rather than being applied consistently.
A. Oduroye, Temilade Opanuga, Esther Tosin Akanbi et al.· International journal of re...· 0 citations