Research Article
Empirical Evaluation of Noise Influence on Supervised Machine Learning Algorithms Using Intrusion Detection Datasets
Table 6
Comparison between different datasets related to intrusion detection.
| No. | Dataset | Creation date | Based on | Deployment | Attack categories | No. of attributes | No. of records |
| 1 | DARPA 1998 | 1998 | — | NIDS | DoS | 41 | Not mentioned | Probe | R2L | U2R |
| 2 | KDD CUP 99 | 1999 | DARPA 1998 | NIDS | DoS | 41 | 4,898,431 | Probe | R2L | U2R |
| 3 | NSL-KDD | 2009 | KDD CUP 99 | NIDS | DoS | 41 | 148,517 | Probe | R2L | U2R |
| 4 | UNSW-NB15 | 2015 | — | NIDS | Analysis Reconnaissance | 45 | 2,540,044 | Shellcode | Fuzzers | Worm | Generic | DoS | Exploits | Backdoors |
| 5 | ADFA | 2013 | — | HIDS | Hydra-FTP | Not mentioned | 5,773 (Windows) | Hydra-SSH | 5,800 (Linux) | Add-user | Java-Meterpreter | Web-shell |
|
|