Innovative Machine Learning Approaches for Early Diabetes Prediction: Enhancing Accuracy and Timeliness in Disease Detection through Advanced Predictive Analytics

S. Prasanna, S. Karimulla, K. Yella Reddy, K. Tharakananda, P. Sateesh Kumar, G. Suneel

2025

Abstract

Detection at an early stage is important so that diabetes can be managed properly and severe complications like heart disease, stroke, renal failure, and vision loss are avoided complication about early detection. Currently, more than 420 million people in the world have diabetes, and it is expected that this prevalence will continue to increase; hence there is a need for new predictive methods. Traditional diagnosis depends on biochemical tests and clinical assessments that do not always reveal early warning signs. However, advanced machine learning techniques present exciting alternatives by sifting through massive amounts of complex data to find early signs of diabetes development. This study surveys multiple machine learning models: random forests, logistic regression-nearest neighbors, decision trees, gradient boosting, LGBM deep learning networks, hybrid models. Each model introduces unique strengths; random forests are highly robust, gradient boosting predictive performance is maximally enhanced plus deeplearning networks are particularly proficient in pattern recognition. Hybrid and ensemble methods are basically multiple models for higher accuracy reinforcement learning can also adapt itself with changing data patterns. This research will use all these diverse machine learning techniques to improve diabetic prediction's accuracy as well as speed to eventually enhance patient outcomes along with public health strategies.

Paper Citation

in Harvard Style

Prasanna S., Karimulla S., Reddy K., Tharakananda K., Kumar P. and Suneel G. (2025). Innovative Machine Learning Approaches for Early Diabetes Prediction: Enhancing Accuracy and Timeliness in Disease Detection through Advanced Predictive Analytics. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 840-845. DOI: 10.5220/0013890700004919

in Bibtex Style

@conference{icrdicct`2525,
author={S. Prasanna and S. Karimulla and K. Reddy and K. Tharakananda and P. Kumar and G. Suneel},
title={Innovative Machine Learning Approaches for Early Diabetes Prediction: Enhancing Accuracy and Timeliness in Disease Detection through Advanced Predictive Analytics},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={840-845},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013890700004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - Innovative Machine Learning Approaches for Early Diabetes Prediction: Enhancing Accuracy and Timeliness in Disease Detection through Advanced Predictive Analytics
SN - 978-989-758-777-1
AU - Prasanna S.
AU - Karimulla S.
AU - Reddy K.
AU - Tharakananda K.
AU - Kumar P.
AU - Suneel G.
PY - 2025
SP - 840
EP - 845
DO - 10.5220/0013890700004919
PB - SciTePress