“Dimension Reduction: Feature Subset” Method for Selecting the Best Index Combination in Reputation Evaluation of Crowdsourcing Participants

Table 3

Accuracy of ten-fold cross-validation of the reputation evaluation index selection classifier.

Model	DT	SVM	BPNN	RBFNN	NB	KNN	Rank
ReliefF	0.899	0.901	0.910	0.856	0.878	0.883	3.67
MIV	0.875	0.887	0.886	0.828	0.819	0.872	1.67
PCA	0.817	0.844	0.855	0.849	0.846	0.854	1.5
LDA	0.881	0.908	0.905	0.848	0.871	0.893	3.17

The best result for each classifier is in bold.