Figure 1. In the Lasso model, a five-fold cross-validation approach was used for the choice of optimal parameters.

(A) In the Lasso model, a five-fold cross-validation approach was used for the choice of optimal parameters.Using the partial likelihood anomaly curve and the log (lambda) plot, the vertical line was drawn at the optimal value to obtain the included feature factors. (B) The lambda curve generated a profile based on the log (lambda) sequence. Vertical lines were drawn at the values selected using the five-fold cross-validation method, with 20 characteristic factors being selected. (C) The algorithm of SVM-RFE support vector machine was used to further screen the 20 characteristic factors. Finally, a prediction model with 9 best features with an average 10-fold cross-validation score of 0.8914 was established.