Table 23.

Performance comparison of our method vs. CMIM for the balanced E. coli data set when identical numbers of features are selected.

	AUC		Precision		Recall		F1		MCC
N4	0.780	0.769	0.733	0.719	0.701	0.696	0.717	0.707	0.446	0.424
N5	0.779	0.771	0.730	0.715	0.706	0.696	0.718	0.705	0.445	0.419
N6	0.762 <	0.771	0.735	0.716	0.663	0.684	0.696	0.699	0.425	0.413
N7	0.783	0.769	0.737	0.711	0.696	0.696	0.716	0.703	0.448	0.413
N8	0.781	0.767	0.723	0.709	0.711	0.697	0.717	0.703	0.439	0.412
N9	0.782	0.767	0.715	0.720	0.703	0.700	0.709	0.710	0.423	0.421
N10	0.781	0.767	0.725	0.705	0.702	0.702	0.713	0.704	0.436	0.409
N11	0.777	0.765	0.719	0.706	0.700	0.700	0.709	0.703	0.426	0.408
N12	0.776	0.764	0.715	0.703	0.695	0.698	0.705	0.700	0.418	0.404
N13	0.776	0.765	0.731	0.704	0.695	0.700	0.712	0.702	0.439	0.406