• anomaly detection
• deep neural networks
• intrusion detection system

This thesis proposes an inspection of binary and multiclass classification in the case of an anomaly detection-based network intrusion detection system through the use of five different deep neural networks. First of all, a brief description of which is an intrusion detection system, a neural network and what are the types of network used. Then the testbed on which all the performance results are calculated, is presented. The last step is to evaluate the results comparing the performances in the case of binary and multiclass classification.