KNN Optimization Using Grid Search Algorithm for Preeclampsia Imbalance Class

Abstract

The performance of predicted models is greatly affected when the dataset is highly imbalanced and the sample size increases. Imbalanced training data have a major negative impact on performance. Currently, machine learning algorithms continue to be developed so that they can be optimized using various methods to produce the model with the best performance. One way of optimization with apply hyperparameter tuning. In classification, most of the algorithms have hyperparameters. One of the popular hyperparameter methodologies is Grid Search. GridSearch using Cross Validation makes it easy to test each model parameter without having to do manual validation one by one. In this study, we will use a method in hyperparameter optimization, namely Grid Search. The purpose of this study is to find out the best optimization of hyperparameters for two machine learning classification algorithms that are widely used to handle imbalanced data cases. Validation of the experimental results uses the mean cross-validation measurement metric. The experimental results show that the KNN model gets the best value compared to the Decision Tree.