Ensemble Learning Boosting Model of Improving Classification and Predicting

Abstract

Artificial Intelligence Engineering is an important topic and has been studied extensively in various
fields. Machine learning is part of Artificial Intelligence that has been used to solve prediction
problems and financial decision making. An effective prediction model is one that can provide a
higher prediction accurate, that is the goal of prediction model development. In the previous
literature, various classification techniques have been developed and studied, which by combining
several classifier approaches have shown performance over a single classifier. In building a
boosting ensemble model, there are three critical issues that can affect model performance. First
are the classification techniques actually used; the second is a combination method for combining
several classifiers; and all three classifiers to be combined. This paper conducts a comprehensive
study comparing the ensemble boosting classifier and three widely used classification techniques
including AdaBoost, Gradient boosting, XGB Classifier. The results of the experiment with two
financial ratio datasets show that the Ensemble Boosting Classifier has the best performance with
an accurate value of 98%, while AdaBoost is 96%, Gradient_boosting is 98%, and XGB Classifier
is 98%. Ensemble Boosting matches all available data, so the predict () function can be called to
make predictions on new data.