The core goal of this thesis is to employ this strategy to ensure the robustness of the proposed model. For all measurements, no single classifier technique exceeded the others. However, when examining balanced datasets against imbalanced datasets, the combination of various machine learning algorithms and application of the voting technique resulted in measures that exceeded the individual machine learning techniques. F1-score measurements show that the proposed model performs better than the individual models. Furthermore, the results obtained with the IBM imbalanced dataset are statistically significant.