Ensemble Learning method for improving the Healthcare IoT System

Abstract

Wireless Body Area Network (WBAN) is several wearable sensor nodes with unstable sensing, storage, computation, and communication capabilities. Heart infection is an essential origin of death internationally, and early recognition is vital in avoiding the development of the infection. This article presents an Ensemble Learning (EL) method for improving the Healthcare Internet of Things (IoT) System. This approach aims to forecast the risk of heart infection. This method involves dividing the dataset into subgroups at random using a mean-based split. The data is then divided into multiple groups using a classification and regression tree. A weighted aging classifier ensemble is used to create an ensemble from the many trees used for classification and regression. It makes sure an optimal function is reached. The experimental outcomes on the datasets reached better classification accuracies that outperformed other machine learning algorithms like Artificial Neural Network, Naıve Bayes, and Support Vector Machine. The outcomes demonstrate that the risk of heart infection can be forecasted efficiently through ensemble learning.