Achieving optimal performance of machine learning models is a critical challenge, particularly when dealing with biased datasets. This issue is particularly prevalent in imbalanced datasets, where classifiers prioritize the majority class while overlooking the minority class. In this work, we employ fundamental balancing strategies to mitigate data imbalance and improve model performance. To test the effects of data balancing techniques on the Machine Learning algorithm’s performance, four classifiers namely: C4.5, Random Forest, Multilayer Perceptron, and Naïve Bayes are adopted. Accordingly, this study focuses on the effectiveness of a new simple approach, which handles effectively the problem of data imbalanced distribution, and avoids some drawbacks of other balancing techniques addressed in the literature.