Enhancing Diabetes Detection: A Weighted Averaging Approach for Combined Model Accuracy

Diabetes, marked by prolonged high blood sugar levels, poses a significant global health challenge. Precise early prediction is vital but faces hurdles due to limited data and complexities like outliers. Uncontrolled diabetes heightens severe complications, driving research into advanced machine and deep learning techniques. Healthcare professionals recognize the crucial role of Machine Learning Algorithms (MLA) in enhancing disease identification. This study integrates deep and machine learning models through techniques like K-means, CNN, NB, and SVM. Combining their predictions using weighted averaging yields a 79.5% accuracy, improving diabetes prediction globally.