Prediction of Diabetes Using Machine Learning Approach

Diabetes is a chronic disease posing a significant global health challenge, with the number of affected individuals projected to nearly double by 2045. This disease, which is characterized by high blood sugar, can cause severe complications such as kidney failure, blindness, and heart problems. To address this pressing concern this study target to enhanced machine learning technology-based system for the early understanding of diabetes with high accuracy. The suggested method makes use of the output of different technique, such as Decision Tree, Random Forest, K-Nearest Neighbor and Naive Bayes, to maximize their benefits and enhance overall prediction performance. The study evaluates the accuracy of each algorithm individually and selects the techniques with the best result as the final predictive system. The results of this study support ongoing initiatives to apply innovative data science methods for diabetes early diagnosis and treatment, ultimately assisting in the avoidance of the disease's severe consequences and enhancing patient outcomes. Based on the Random Forest classifier model, this model generated an accuracy of 96.83%.