A Novel SMOTE-Based Classification Approach to Online Data Imbalance Problem

Table 5

Comparative results on Macao forecasting data in 2011.

Algorithms	DGSMOTE	OS-ELM	MCOS-ELM	LS-SVM
Testing time (s)	0.0321	0.0412	0.0425	1.638
Minority training accuracy	0.9469	0.5723	0.7206	0.7457
Majority training accuracy	0.9456	0.9756	0.9673	0.9492
Whole testing accuracy	0.9495	0.9421	0.9662	0.9263
-mean	0.9347	0.7387	0.8399	0.7948