Research Article

Detecting Unknown Threat Based on Continuous-Time Dynamic Heterogeneous Graph Network

Table 3

Detection result on different number of the abnormal samples of CDHGN.

Dataset divisionAbnormal sampleAUC (test)
#Train#Val#Test

0.27 : 0.03 : 0.7187214830.9366
0.22 : 0.04 : 0.74152285110.9400
0.258 : 0.002 : 0.7417915110.9400