Unite and conquer with bagging-boosting “UC2B” for enhanced anomaly detection

The increasing importance of cybersecurity de-mands effective methods for detecting anomalous behavior in computer systems and networks. This paper presents an approach for anomaly detection that combines Unite and Conquer techniques with Bagging and Boosting methods based on machine learning and statistical models. The proposed approach was evaluated using the UNSW-NB15 dataset, which is a bench-mark for cybersecurity evaluation. The results demonstrate the effectiveness of the combination of Unite and Conquer approaches with Bagging and Boosting in detecting anomalies in cybersecurity. This research provides a valuable contribution to anomaly detection, achieving an accuracy of up to 99.5 % on unseen data. This suggests that the proposed approach may have promising potential for practical applications, especially in real-world scenarios where high computing power is not a prerequisite.