MAE-CAD: An IP-Based Core Network Asset Discovery Technology Based on Multiple Autoencoders

Table 7

Detection results of MAE-CAD trained with the training set of different proportions of core-network-assets (positive samples): noncore-network-assets (negative samples) during the fine-tuning phase.

Positive: Negative	Acc (%)	P (%)	R (%)	F1 (%)
1 : 1	95.74	94.55	95.54	95.04
1 : 4	93.11	92.79	90.91	91.84
1 : 8	93.41	92.33	92.22	92.28
1 : 24	92.95	91.82	91.66	91.74
1 : 50	93.04	92.8	90.73	91.75
1 : 100	93.02	92.46	91.08	91.76
1 : 200	92.91	92.99	90.20	91.57