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

Table 2

Comparative results on chessboard-shaped dataset.

Algorithms	DGSMOTE	OS-ELM	MCOS-ELM	LS-SVM
Testing time (s)	0.0874	0.1976	0.0531	132.476
Minority training accuracy	0.9167	0.6329	0.7438	0.6385
Majority training accuracy	0.9414	0.8988	0.8051	0.9792
Whole testing accuracy	0.9363	0.8439	0.8612	0.9606
-mean	0.9325	0.7334	0.7445	0.7822