Binary text classification using genetic programming with crossover-based oversampling for imbalanced datasets

It is well known that classifiers trained using imbalanced datasets usually have a bias toward the majority class. In this context, classification models can present high performance overall and for class, even when minority class significantly lower. This paper presents genetic programming (GP) model with crossover-based oversampling technique dataset binary text classification. The aim of study to apply an solve issue improve GP employed proposed technique. employs crossover operator generating new samples in dataset. By combination GP, was improved. shown outperforms all applications use original without resampling. Moreover, system surpassed approaches synthetic (SMOTE) random oversampling. Further comparison state-of-the-art on five terms F1-score shows superior approach.