A Feature Selection Algorithm Integrating Maximum Classification Information and Minimum Interaction Feature Dependency Information

Table 4

Average classification accuracy (%) of C4.5 classifier.

Data set	NDCRFS	MIM	IG-RFE	IWFS	CMIM	DWFS	CIFE
Lymphography	43.935	41.893	41.473	41.347	42.322	43.002	42.322
Dermatology	95.021	94.434	94.149	94.187	95.021	93.337	94.727
Cardiotocography	98.401	98.401	98.401	98.401	98.401	98.401	98.401
Pendigits	94.569	94.343	94.196	93.782	93.768	94.222	93.675
Lung	87.774	79.918	85.113	75.964	83.842	84.157	77.236
Carcinom	70.604	54.586	25.79	48.292	56.822	53.999	24.3
Nci9	69.929	61.012	65.095	60.667	71.083	57.929	60.226
PCMAC	87.906	86.464	86.515	82.502	85.897	86.669	80.805
Pixraw10P	99.0	97.0	96.0	92.0	95.0	92.0	95.0
SMK-CAN-187	64.125	62.006	61.494	63.656	62.077	65.747	57.852
Lymphoma	87.75	79.75	80.0	69.528	82.806	69.417	86.917
COIL20	79.876	67.614	72.762	63.186	62.895	70.629	58.295
Average accuracy rate	81.574	76.452	75.082	73.626	77.495	75.792	72.48
Wins/Ties/Losses		11/1/0	11/1/0	11/1/0	10/1/1	10/1/1	11/1/0