Detecting web attacks using random undersampling and ensemble learners

Journal of Big Data

Table 11 Results for CatBoost

Sampling	AUC	SD AUC	AUPRC	SD AUPRC
None	0.62759	0.0195	0.322	0.05891
999:1	0.64623	0.02485	0.34685	0.05637
99:1	0.75901	0.03296	0.37749	0.06585
95:5	0.7931	0.01928	0.30189	0.096
9:1	0.80603	0.02243	0.17467	0.08558
3:1	0.89987	0.02741	0.14613	0.10772
65:35	0.92978	0.01256	0.13726	0.09471
1:1	0.9318	0.00712	0.08926	0.08841

Sampling column reports negative to positive class ratio after applying random undersampling, or “None” for case when no undersampling is applied
AUC stands for Area Under the Receiver Operating Characteristic Curve, SD stands for standard deviation reported for every metric, AUPRC stands for Area under the Precision-Recall Curve