A Novel Ensemble Credit Scoring Model Based on Extreme Learning Machine and Generalized Fuzzy Soft Sets

Table 4

Results for all single classifiers without AEnet-based feature selection.

Data set	Performance measures	Single classifiers
		C5.0	SVM-R	DBN	Bayes	ELM
Germany	AUC	0.680	0.732	0.753	0.761	0.764
	HM	0.336	0.362	0.399	0.338	0.379
	BS	0.251	0.167	0.173	0.199	0.172
	ACC	0.745	0.752	0.749	0.745	0.762
Australia	AUC	0.867	0.812	0.876	0.857	0.852
	HM	0.516	0.603	0.613	0.581	0.610
	BS	0.148	0.113	0.107	0.166	0.127
	ACC	0.736	0.742	0.751	0.734	0.735
Japan	AUC	0.857	0.908	0.911	0.888	0.882
	HM	0.689	0.608	0.608	0.667	0.618
	BS	0.156	0.120	0.138	0.176	0.142
	ACC	0.738	0.749	0.757	0.742	0.751
Iran	AUC	0.616	0.604	0.614	0.715	0.710
	HM	0.111	0.074	0.061	0.192	0.133
	BS	0.071	0.071	0.078	0.077	0.083
	ACC	0.623	0.617	0.640	0.627	0.648
Bene 1	AUC	0.728	0.821	0.767	0.740	0.752
	HM	0.310	0.366	0.250	0.335	0.351
	BS	0.267	0.176	0.168	0.297	0.199
	ACC	0.701	0.749	0.698	0.690	0.753
Bene 2	AUC	0.745	0.789	0.746	0.707	0.762
	HM	0.385	0.309	0.238	0.276	0.270
	BS	0.143	0.178	0.142	0.172	0.153
	ACC	0.728	0.720	0.715	0.711	0.747
Shuttle	AUC	0.604	0.675	0.693	0.714	0.722
	HM	0.299	0.334	0.367	0.317	0.358
	BS	0.224	0.153	0.158	0.186	0.163
	ACC	0.662	0.693	0.740	0.698	0.720
Skin_segment	AUC	0.770	0.749	0.804	0.802	0.805
	HM	0.458	0.556	0.564	0.543	0.576
	BS	0.131	0.105	0.099	0.156	0.119
	ACC	0.653	0.685	0.689	0.687	0.695
MiniBooNE	AUC	0.761	0.837	0.838	0.833	0.833
	HM	0.613	0.560	0.559	0.625	0.584
	BS	0.140	0.112	0.126	0.164	0.134
	ACC	0.655	0.690	0.696	0.695	0.710
LC2017Q1	AUC	0.547	0.557	0.563	0.669	0.671
	HM	0.099	0.067	0.057	0.181	0.126
	BS	0.062	0.066	0.072	0.071	0.078
	ACC	0.655	0.670	0.688	0.687	0.712