Constructing Better Classifier Ensemble Based on Weighted Accuracy and Diversity Measure

Table 4

The average size of subensembles selected by candidate measures and FHC.

Dataset	WAD	Kappa-Error	GenDiv	Acc-Only	Div-Only	No-Selection
Hepatitis	27	5	31	9	59	200
Autos	21	12	36	14	43	200
Heart-statlog	20	23	56	11	54	200
Ionosphere	17	4	53	10	65	200
Credit-a	27	3	40	15	34	200
Diabetes	25	9	26	3	66	200
Vehicle	29	18	29	13	56	200
Car	18	11	19	9	69	200
cmc	22	18	43	2	45	200
Segment	24	10	109	4	43	200
kr-vs-kp	25	9	69	9	12	200
Hypothyroid	23	25	33	12	34	200
Waveform	15	10	30	27	78	200
Page-blocks	29	16	37	6	45	200
Nursery	28	7	36	4	78	200
Average	23.3	12	43.1	9.9	52.1	200