Improving the reliability of network intrusion detection Systems through data set integration

MCA

Abstract— The suggested IDS paradigm is designed to identify network intrusions by categorising every network packet that crosses it as benign or malicious. The dataset from the Canadian Institute for Cyber Security Intrusion Detection System (CICIDS2017) was utilised to train and evaluate the model that was suggested. The model was tested in terms of its total precision, alerting rate, false alarm rate, and training overhead.DDOS assaults were trained and validated using the Canadian Institute for Cyber Security Intrusion Detection System (KDD Cup 99) database. We are compared to two datasets (CICIDS2017 and KDD Cup 99). The Deep learning algorithms must then be implemented as Proposed Method Classification Using LSTM technique Model predict.Finally, testing dataset for anomaly detection model The Deep learning algorithms must then be implemented as Proposed Method Classification Using LSTM algorithm Model predict.The testing dataset for the anomaly detection model was eventually classed as attack or normal. Finally, the experimental findings suggest that performance measurements including accuracy, precision, recall, and confusion matrix are effective.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.