Protein Sequence Classification with Improved Extreme Learning Machine Algorithms

Table 1

Successful testing classification rate and standard derivation (rate (%) ± Dev (%)) comparisons of different classifiers: fixed protein sequence training dataset (pir1) and testing dataset (pir2), where L, S, and G stand for the linear kernel, the sigmoid kernel, and the Gaussian kernel, respectively.

Methods	𝐋	𝐒	𝐆
VOP-ELM	92.51 ± 0	92.64 ± 1.6	93.69 ± 0.89
V-ELM	87.45 ± 0.01	90.14 ± 0.71	90.46 ± 0.69
OP-ELM	91.39 ± 0.01	91.51 ± 1.0	92.28 ± 0.70
SVM	90.63 ± 0.01	90.63 ± 0.01	90.63 ± 0.01
BP	85.77 ± 0.42	88.39 ± 1.69	88.77 ± 1.35
ELM	85.39 ± 0	87.66 ± 1.2	87.80 ± 1.4