The CICIDS dataset is one of the most important and it is crucial to understand the intrusion detection which occurs in the area of cybersecurity research. It identifies many detect and it can be classified into two broad classes such as benign and malicious network traffic that occurs frequently in network-based environment. Such network traffic or anomaly can be easily identified with the help of promising computing area such as machine learning and deep learning. 12 types of attacks are focused on this research and trained using various machine learning algorithms. In order to accurately classify a stacked classifier of various boosting techniques are combined and using a meta classifier an accurate classification of defects has been identified which yielded an accuracy of 95%.