iMPT-FDNPL: Identification of Membrane Protein Types with Functional Domains and a Natural Language Processing Approach

Table 3

Performance of different multilabel classifiers with features derived from functional domain information via one-hot scheme.

Scheme (base classifier)	Exact match	Accuracy	Hamming loss	Integrated score
RAKEL (RF)	0.825	0.827	0.061	0.6406
RAKEL (SVM-polynomial kernel)	0.847	0.848	0.054	0.6794
RAKEL (SVM-RBF kernel)	0.846	0.847	0.054	0.6778
BR (RF)	0.785	0.788	0.049	0.5882
BR (SVM-polynomial kernel)	0.774	0.778	0.049	0.5726
BR (SVM-RBF kernel)	0.836	0.840	0.048	0.6685