This research introduces a machine learning framework for early **diabetes prediction** using a **voting classifier approach**. Logistic Regression is compared against Decision Trees, SVM, and Random Forest using a historical dataset with health attributes including **glucose levels, BMI, blood pressure**, and **family history**. Logistic Regression demonstrated **91% accuracy**, outperforming others and showcasing reliability and computational efficiency for healthcare analytics.
Keywords
Diabetes Prediction, Machine Learning, Logistic Regression, Binary Classification, Medical Diagnosis
Conclusion
The voting classifier-based diabetes prediction system offers a **robust, interpretable**, and highly accurate method for medical diagnostics. Its modular design supports **real-time prediction**, seamless deployment in mobile/web platforms, and scalability across healthcare domains. The model’s performance and usability underline its suitability for early diagnosis in clinics, camps, and telemedicine applications—contributing to improved **preventive care and patient outcomes**.
References
Reddy & Ramakrishnan (2022). E2E-DPV: Ensemble Voting for Diabetes Prediction. Journal of Medical Systems.
Post Comment