Improving Diabetes Prediction Accuracy Using Optimized Machine Learning Techniques
Ms. Simran Shinde
, Pranav Hanumant Shiralkar
Diabetes Prediction, Machine Learning, PIMA Indians Dataset, Random Forest, Classification, Healthcare Analytics, False Negative Reduction
Diabetes is a widespread chronic disease where early detection is critical to preventing serious complications. Traditional diagnostic methods depend on laboratory tests that can delay risk identification. This study improves diabetes prediction accuracy using optimized machine learning techniques applied to the PIMA Indians Diabetes Dataset — containing physiological indicators including glucose level, blood pressure, BMI, insulin, and age. Preprocessing (missing value handling, feature scaling, normalization) is applied to ensure data quality. A supervised classification model is developed and evaluated using accuracy, precision, recall, and confusion matrix analysis, with emphasis on minimizing false negatives. Results confirm that systematic preprocessing and feature analysis significantly improve classification reliability, supporting early screening and data-driven clinical decisions.
"Improving Diabetes Prediction Accuracy Using Optimized Machine Learning Techniques", JETNR - JOURNAL OF EMERGING TRENDS AND NOVEL RESEARCH (www.JETNR.org), ISSN:2984-9276, Vol.4, Issue 4, page no.b312-b316, April-2026, Available :https://rjpn.org/JETNR/papers/JETNR2604166.pdf
Volume 4
Issue 4,
April-2026
Pages : b312-b316
Paper Reg. ID: JETNR_233340
Published Paper Id: JETNR2604166
Downloads: 00034
Research Area: Science and Technology
Country: -, -, India
ISSN: 2984-9276 | IMPACT FACTOR: 9.87 Calculated By Google Scholar | ESTD YEAR: 2023
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.87 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: RJPN (IJPublication) Janvi Wave
