Comparative Study of SMOTE and Bootstrapping Performance Based on Predication Methods

Recently, there has been a renewed interest in smart health systems that aim to deliver high quality healthcare services. Prediction methods are very essential to support these systems. They mainly rely on datasets with assumptions that match the reality. However, one of the greatest challenges to prediction methods is to have datasets which are normally distributed. This paper presents an experimental work to implement SMOTE (Synthetic Minority Oversampling Technique) and bootstrapping methods to normalize datasets. It also measured the impact of both methods in the performance of different prediction methods such as Support vector machine (SVM), Naive Bayes, and neural network(NN) The results showed that bootstrapping with native bays yielded better prediction performance as compared to other prediction methods with SMOTE.