Research Article
iMPT-FDNPL: Identification of Membrane Protein Types with Functional Domains and a Natural Language Processing Approach
Figure 4
Boxplot to show the performance of classifiers with different feature types using tenfold cross-validation for 20 times. (a) Accuracy; (b) exact match; (c) hamming loss; (d) integrated score. Features derived from functional domain via word2vector are most efficient to identify membrane protein types.
(a) |
(b) |
(c) |
(d) |