Table 4: Comparison of prediction results for 2015 dataset based on OC-SVM and IF combined SVM classifiers at different updating ratios.

Updating rateOne-class SVM (OC-SVM)Isolation forest (IF)
Recall 1Recall 2MARaRecall 1Recall 2MAR

10%0.82750.73240.7799 G0.84140.72570.7835 G
20%0.84820.83860.8434 E0.86170.79830.8300 F
30%0.86120.87490.8680 D0.87170.83120.8515 E
40%0.87390.90770.8908 B0.87670.85990.8683 D
50%0.88240.92380.9031 A0.88020.87270.8765 C,D
60%0.88670.93790.9123 A0.88600.87700.8815 B,C

aMAR: macro average recall. Means with the same letter(s) are not significantly different at 0.01 level.