Predicting Cross-Species Infection of Swine Influenza Virus with Representation Learning of Amino Acid Features

Figure 2

Feature representation optimization with the mRMR algorithm. (a) The SFS curves for ACC of “class” and “prob” feature. The feature number (1-64) and the accuracy were represented by the - and -axis. (b) The SFS curves for MCC of “class” and “prob” feature. The feature number (1-64) and the coefficient were represented by the - and -axis. The “class” and “prob” features were marked by the blue and yellow color.

(a)

(b)