Skip to main content
. 2020 Apr 15;8:e8793. doi: 10.7717/peerj.8793

Figure 1. Demographic and clinical feature selection using the LASSO binary logistic regression model and support vector machine algorithm.

Figure 1

(A) In the Lasso model, the choice of the optimal parameters used a five-fold cross-validation approach. Using the partial likelihood anomaly curve and the log (lambda) plot, the vertical line is drawn at the optimal value to obtain the included feature factors. (B) The lambda curve generates a profile based on the log (lambda) sequence. Vertical lines were drawn at the values selected by the five-fold cross-validation method, and 16 characteristic factors were selected. (C) Using the support vector machine SVM-RFE algorithm to further screen these 16 characteristic factors, we finally established a prediction model with eight best features with an average 10-fold cross-validation score of 0.8531.