Figure 2.

(A) Feature selection via the least absolute shrinkage and selection operator (LASSO) regression model. (A) The LASSO coefficient profiles of 19 features. The coefficient profile plot was conducted against the log (lambda, λ) sequence. The dotted vertical line was drawn at the lambda with a minimum mean squared error (lambda.min); nine features were selected by the LASSO regression. The solid vertical line was drawn at the lambda.min with one standard error (lambda.1se); four features were selected by the LASSO regression model. (B) Feature selection via 10-fold cross-validation. (B) The optimal parameter (lambda, λ) selection in the Lasso regression model used 10-fold cross-validation via the minimum criteria. The partial likelihood deviance (binomial deviance) curve was plotted versus logλ. The dotted vertical line was drawn at the lambda with a minimum mean squared error (lambda.min); nine features were selected. The solid vertical line was drawn at the lambda.min with one standard error (lambda.1se); four features were selected.