Application of an interpretable classification model on Early Folding Residues during protein folding

Table 3 Parameter selection to obtain the results of Table 2 using the Weka plug-in

Parameter	Run 1	Run 2	Run 3	Run 4	Run 5	Run 6
Cost function to optimize	CA	WCA	F₁	CA	WCA	F₁
Number of epochs	150	150	150	250	250	250
Number of prototypes	1	1	1	5	5	5
Data point ratio per round	0.75	0.75	0.75	0.75	0.75	0.75
Sigmoid sigma interval	[1.0,5.0]	[1.0,15.0]	[1.0,50.0]	[1.0,5.0]	[1.0,15.0]	[1.0,50.0]
Prototype learning rate	1.0	1.0	1.0	1.0	1.0	1.0
Matrix learning	True	True	True	True	True	True
Omega learning rate	1.0	1.0	1.0	1.0	1.0	1.0
Omega dimension	27	27	27	27	27	27
Cost function beta	-	-	1	-	-	1
Cost function weights	-	[0.75,0.25]	-	-	[0.75,0.25]	-
Parallel execution	True	True	True	True	True	True

Classification accuracy (CA), weighted classification accuracy (WCA) with weights 0.75 and 0.25 as well as F _β-measure with β=1 (F₁)

ISSN: 1756-0381