Cybercrime: Identification and Prediction Using Machine Learning Techniques

Table 3

Performance metrics using the SVM classifier.

Attribute	TP	TN	FP	FN	FAR	DR	ACC	Precision	Recall	Specificity	Sensitivity	FMI
1	16	64	20	0	0.2381	1	0.8	0.23810	0.2	1	0.7619	0.6667
2	16	64	20	0	0.2381	1	0.8	0.23810	0.2	1	0.7619	0.6667
3	80	0	9	11	1.00	0.879	0.8	2.22222	0.2	0.8791	0.00	0.8889
4	80	0	16	4	1.00	0.952	0.8	1.25000	0.2	0.9523	0.00	0.8909
5	80	0	0	20	0.00	0.8	0.8	0.00000	0.2	0.8	0.00	0.8944
6	80	0	0	20	0.00	0.8	0.8	0.00000	0.2	0.8	0.00	0.8944
7	80	0	0	20	0.00	0.8	0.8	0.00000	0.2	0.8	0.00	0.8944
8	80	0	0	20	0.00	0.8	0.8	0.00000	0.2	0.8	0.00	0.8944
9	33	47	20	0	0.298	1	0.8	0.29851	0.2	1	0.7014	0.7891
10	80	0	0	20	0.00	0.8	0.8	0.00000	0.2	0.8	0.0000	0.8944