57
Annex C: list of all the classification reports
Below, there are two lists, containing all the classification reports that resulted from the
analysis of the classification models in the two different use cases. Further information about
the models and discussion of their performance can be found in section Classification models
analysis, as well as information about the interpretation of the classification reports.
C.1. Case A: using KDDTrain+ KDDTest+ as training and test sets
Logistic Regression:
Table 11: logistic regression on the multiclass training set
precision recall f1-score support
back
0.99
0.97
0.98
974
buffer_overflow 0.60
0.86
0.71
21
ftp_write
0.25
1.00
0.40
2
guess_passwd
0.98
0.96
0.97
54
imap
0.91
1.00
0.95
10
ipsweep
0.97
0.97
0.97
3617
land
0.61
0.85
0.71
13
loadmodule
0.22
0.67
0.33
3
multihop
0.29
0.40
0.33
5
neptune
1.00
1.00
1.00
41223
nmap
0.96
0.93
0.94
1540
normal
1.00
0.99
0.99
67470
perl
0.00
0.00
0.00
0
phf
1.00
1.00
1.00
4
pod
1.00
1.00
1.00
202
portsweep
0.98
1.00
0.99
2890
rootkit
0.20
0.50
0.29
4
satan
0.95
0.98
0.97
3520
smurf
1.00
0.99
0.99
2678
spy
0.00
0.00
0.00
0
teardrop
1.00
1.00
1.00
891
warezclient
0.82
0.88
0.85
833
warezmaster
0.80
0.84
0.82
19
accuracy
0.99
125973
macro avg
0.72
0.82
0.75
125973
weighted avg
0.99
0.99
0.99
125973