Nasopharyngeal Carcinoma Subtype Discovery via Immune Cell Scores from Tumor Microenvironment

Figure 7

Construction and validation of machine learning models for predicting NPC subtypes. (a) Parameter optimization of RF model by 5-fold cross-validation in the training set. The mtry: number of variables randomly sampled as candidates at each split. (b) ROC curve analysis of the performance of RF in the test set. (c) Parameter optimization of SVM model by 5-fold cross-validation in the training set. C parameter in SVM is penalty parameter of the error term. (d) ROC curve analysis of the performance of RF in the test set.

(a)

(b)

(c)

(d)