Smart Healthcare System for Disease Prediction using Hybrid Machine Learning

Accurate disease prediction is essential in healthcare systems to enable early identification and intervention for better patient outcomes. This paper provides a hybrid strategy for predicting diseases and assesses its effectiveness with and without hyperparameter adjustment. The outcomes show that the hybrid technique can predict the presence or absence of disease with an accuracy of 94.3% without hyperparameter adjustment. However, the performance of the hybrid technique significantly improves with hyperparameter optimization, achieving a remarkable accuracy of 99.1%. These results highlight the significance of hyperparameter tuning in machine learning model optimization and its potential to improve illness prediction systems' performance and accuracy. With its high accuracy and capacity for complicated pattern adaptation, the proposed hybrid technique shows potential for improving disease prediction in existing healthcare systems.