Online Feature Selection (OFS) with Accelerated Bat Algorithm (ABA) and Ensemble Incremental Deep Multiple Layer Perceptron (EIDMLP) for big data streams

Journal of Big Data

Table 2 Classification results of Arcene

Classifiers	FS algorithms	Accuracy	Precision	Recall	F-measure	Processing time in seconds
NB	FS-PSO	81	80.7	80.6	80.68	58.9
	FS-APSO	81	80.76	80.84	80.77	5.32
	FS-ASAMO	82	82.08	81.25	81.66	0.15
	MR-OFS-ABA	83	82.98	82.38	82.68	0.07
SVM	FS-PSO	85	84.84	84.65	84.75	0.05
	FS-APSO	86	86.29	86.76	86.5	14.89
	FS-ASAMO	87	86.76	86.93	86.84	1.25
	MR-OFS-ABA	87	86.77	87.17	86.97	0.14
HT	FS-PSO	89	88.92	88.71	88.82	0.06
	FS-APSO	90	89.82	89.36	89.88	0.05
	FS-ASAMO	90	90.41	90.34	90.08	0.061
	MR-OFS-ABA	91	90.8	90.99	90.89	0.048
FMCCSC-KNN	FS-PSO	94	93.9	93.91	93.9	0.128
	FS-APSO	95	94.84	95.04	94.94	0.057
	FS-ASAMO	96	95.83	98.69	95.96	0.052
	MR-OFS-ABA	96	96.22	96.42	96.13	0.045
EIDMLP	FS-PSO	99	98.88	98.86	98.9	1.68
	FS-APSO	99	98.8	98.86	98	0.13
	FS-ASAMO	99	99	99.1	98.99	0.062
	MR-OFS-ABA	99	99.12	99.1	98.9	0.053