Ensemble-Based Neighborhood Attribute Reduction: A Multigranularity View

Table 5

Comparisons among the classification accuracies with using NEC.

ID	RAW	APP	MGAPP	RAW	ENT	MGENT	RAW	DIS	MGDIS	RAW	ERR	MGERR
1	0.9232	0.9204	0.9401	0.9232	0.9337	0.9489	0.9232	0.9232	0.9253	0.9232	0.9211	0.9253
2	0.5038	0.5104	0.5125	0.5038	0.5389	0.5600	0.5038	0.3748	0.5204	0.5038	0.5131	0.5176
3	0.9052	0.9037	0.9019	0.9052	0.9211	0.9274	0.9052	0.9052	0.9133	0.9052	0.9107	0.9111
4	0.9557	0.9181	0.9373	0.9557	0.9339	0.9561	0.9557	0.9406	0.9470	0.9557	0.9432	0.9575
5	0.8380	0.8300	0.8440	0.8380	0.8280	0.8400	0.8380	0.8380	0.8320	0.8380	0.8600	0.8600
6	0.7871	0.7794	0.7843	0.7871	0.8030	0.8156	0.7871	0.6843	0.7821	0.7871	0.7965	0.7843
7	0.5454	0.4987	0.5580	0.5454	0.5634	0.5743	0.5454	0.5454	0.5497	0.5454	0.5602	0.5678
8	0.8057	0.8800	0.9057	0.8057	0.8726	0.9154	0.8057	0.8074	0.8646	0.8057	0.8686	0.9023
9	0.7235	0.6729	0.7402	0.7235	0.8235	0.8431	0.7235	0.7235	0.7235	0.7235	0.7644	0.7500
10	0.8089	0.7688	0.8241	0.8089	0.8158	0.8339	0.8089	0.7762	0.8068	0.8089	0.8017	0.8017
11	0.9124	0.8886	0.8905	0.9124	0.8943	0.9038	0.9124	0.9124	0.9171	0.9124	0.9124	0.8990
12	0.6986	0.7090	0.7172	0.6986	0.7120	0.7038	0.6986	0.6985	0.6926	0.6986	0.7134	0.7144
13	0.7639	0.7510	0.7871	0.7639	0.7794	0.7800	0.7639	0.7619	0.7639	0.7639	0.7768	0.6935
14	0.9733	0.9631	0.9811	0.9733	0.9722	0.9656	0.9733	0.9733	0.9756	0.9733	0.9610	0.9675
15	0.5500	0.5246	0.5596	0.5500	0.5500	0.5574	0.5500	0.4779	0.5507	0.5500	0.4469	0.4469
AVG	0.7597	0.7679	0.7922	0.7597	0.7961	0.8084	0.7597	0.7562	0.7843	0.7597	0.7029	0.7799

The greatest values are in bold, and the smallest values are in italics.