Design and Development of an Efficient Network Intrusion Detection System Using Machine Learning Techniques

Today’s internets are made up of nearly half a million different networks. In any network connection, identifying the attacks by their types is difficult task as may have various connections, and number vary from few to hundreds connections. To solve this problem, novel hybrid IDS called NID-Shield proposed in manuscript that classifies dataset according attack types. Furthermore, names found classified individually helping considerably predicting vulnerability individual The NIDS applies efficient feature subset selection technique CAPPER distinct machine learning methods. UNSW-NB15 NSL-KDD datasets utilized for evaluation metrics. Machine algorithms applied training reduced accurate highly merit subsets obtained then assessed cross-validation method attributes. Various performance metrics show with approach achieves good accuracy rate low FPR on shows results when analyzed approaches existing literature studies.