Welcome to ACINT 2023

International Conference on Advanced Computational Intelligence (ACINT 2023)

December 30 ~ 31, 2023, Virtual Conference

Accepted Papers
Train Set Complexity Tuning for Imbalance Learning

Mehmet Ulaş and Mehmet Ali Ergün, Department of Industrial Engineering, İstanbul Technical University, İstanbul, Türkiye


Most of the machine learning algorithms that solve the classification problem give good results assuming that the train set is balanced. Imbalance may frequently occur in real life problems. In this paper, we will develop a solution to make classification algorithms stronger by tuning the complexity of the train set for imbalanced data sets. While adjusting the balance of majority class and minority class in the training set, sample selection will be made by considering the density of the majority class. When we compared the method, we developed with the traditional SMOTE method on a real-life problem, we saw that the method we developed performed better.


Imbalance Learning, Train set complexity tuning, Iterative complexity tuning, Machine Learning.