Enhanced Diabetes Diagnosis Using Ensemble Classifiers with Explainable AI and Oversampling for Imbalanced Data
Bibliographic record
Abstract
Diabetes risk assessment is a key task in healthcare, aiming to identify individuals at risk and initiate early interventions to slow disease progression. Diabetes remains a critical global health concern, necessitating early and accurate diagnostic tools to mitigate long-term complications. This study proposes an enhanced diabetes prediction framework utilizing ensemble machine learning classifiers, explainable AI (XAI), and oversampling technique-SMOTE to address data imbalance. Leveraging the PIMA Indian Diabetes Dataset, the study conducts extensive preprocessing, including outlier handling, skewness correction using Box-Cox transformation, and feature selection via ANOVA F -score. Ensemble classifiers-Random Forest, Extra Trees, Voting, and Stacking were evaluated using evaluation metrics. Among these, the Extra Trees Classifier achieved the highest accuracy of 96.58%, while Random Forest demonstrated strong AUC-ROC performance. Explainable AI technique-SHAP was employed to interpret model predictions and identify key influencing features, such as glucose levels, BMI, and age. The results confirm that ensemble models, when integrated with robust preprocessing and interpretability techniques, significantly enhance the reliability and transparency of diabetes diagnosis. This research highlights that Ensemble classifiers are increasingly being used for diabetes diagnosis, employing machine learning methods to improve accuracy and enable early detection. The integration of oversampling (SMOTE), ensemble learning, and explainable AI significantly improved model performance and interpretability, demonstrating strong potential for clinical decision support systems in diabetes diagnosis.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".