BMC Bioinformatics

Table 3 Machine learning classifiers coupled with encoding methods

From: Protein sequences classification by means of feature extraction with substitution matrices

			Encoding method
Data	Mtr	Clfr	NG	AM	DD	DDSM
DS1	CA	C4.5	96.7	95	95	96.7
		SVM	96.7	93.3	96.7	96.7
		NB	86.7	90	81.7	80
		NN	63.3	78.3	60	61.7
	NA		4935	2060	4905	2565
DS2	CA	C4.5	99.6	99.4	99.8	99.4
		SVM	100	99.4	100	100
		NB	100	74.7	100	100
		NN	100	100	100	98.8
	NA		6503	7055	10058	1312
DS3	CA	C4.5	36.4	-	36.7	79.2
		SVM	43.2	-	43.2	78.94
		NB	43.2	-	43.1	59.4
		NN	20.9	-	21.3	77
	NA		7983	-	8403	508
DS4	CA	C4.5	60	57.5	77.5	82.5
		SVM	67.5	65	87.5	87.5
		NB	57.5	40	92.6	95
		NN	52.5	60	80	80
	NA		5561	3602	7116	5505
DS5	CA	C4.5	75.5	75.1	67.9	73.3
		SVM	84.1	81.2	82.3	82.3
		NB	77.3	63.7	84.5	85.9
		NN	80.5	79.4	78	78
	NA		6465	2393	13830	13083

Mtr: Metric, Clfr: Classifier, CA: Classification Accuracy (%), NA: Number of Attributes.

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com