NSL-KDD

The NSL-KDD data set improves the above problems in the KDD-CUP data set, removes duplicate records in the training set and test set, and increases the proportion of minority samples in the test set, which can better distinguish different models in intrusion detection. Therefore, the NSL-KDD data set is used in this experiment to evaluate the performance of the model.The training set in NSL-KDD is called KDDTrain+ and the test set is called KDDTest+. Each record in the NSL-KDD data set has 42 attributes. 41 attributes represent the characteristic attributes of the data, and 1 attribute represents the type of the attack.