Classifying natural products from plants, fungi or bacteria using the COCONUT database and machine learning

Table 4 Non-default and optimized hyperparameters used in the fivefold cross-validation MAP4, ECFP4, RDKit AP, and properties SVMs comparison

Classifier	Kernel^a	C^a	γ^a
MAP4 SVM	MAP4^b	0.01, 0.1, 1, 10, 100	–
ECFP4 SVM	Tanimoto^c, Dice^c	0.01, 0.1, 1, 10, 100	–
RDKit AP SVM	Tanimoto^c, Dice^c	0.01, 0.1, 1, 10, 100	–
Properties SVM	RBF^d	0.01, 0.1, 1, 10, 100	0.01, 0.1, 1, 10, 100

^aUsed hyperparameters are reported in bold
^bCalculates the similarity matrix between two MAP4 fingerprints, where the similarity of fingerprint a and fingerprint b is calculated (1) counting of elements with the same value and the same index across a and b, and (2) dividing the obtained value by the number of elements of fingerprint a
^cRalaivola et al. [69]
^cRadial basis function (RBF) kernel [70]

ISSN: 1758-2946