Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2012 Aug 27;14(1):87–98. doi: 10.1093/biostatistics/kxs027

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author 2012. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

PMC Copyright notice

Fig. 4. — Classification performances using different methods at the second stage of our approach as well as using different methods directly on the burn patients’ data. Top-left: TPAM, PAM, and PC-PAM; top-right: TLASSO, LASSO, and PC-LASSO; bottom-left: T-ELASTIC NET, ELASTIC NET, and PC-ELASTIC NET; bottom-right: TSVM, SVM, and PC-SVM. ∘: using the Fisher criterion to obtain the direction of projection at the first stage; △: using the first time point; +: using the second time point; ×: combining the early and middle time points; : using the first principal component as the direction of projection at the first stage. Y -axis is the error rate on the test data. The X-axis in the top-left panel is the number of features selected in the models. The X-axes in the top-right and bottom-left panels are the indices of the tuning parameter; each index i (i∈{0,…,99}) corresponds to λ_max−i(λ_max−λ_min)/99; λ_max is the maximum value of the tuning parameter λ calculated by default by R-package glmnet. λ_min=λ_maxλ_min.ratio and λ_min.ratio=0.01, which is the default value of glmnet. The axis in the bottom-right panel is the cost that corresponds to the parameter c in the R-package e1071.

Inline graphic — Classification performances using different methods at the second stage of our approach as well as using different methods directly on the burn patients’ data. Top-left: TPAM, PAM, and PC-PAM; top-right: TLASSO, LASSO, and PC-LASSO; bottom-left: T-ELASTIC NET, ELASTIC NET, and PC-ELASTIC NET; bottom-right: TSVM, SVM, and PC-SVM. ∘: using the Fisher criterion to obtain the direction of projection at the first stage; △: using the first time point; +: using the second time point; ×: combining the early and middle time points; : using the first principal component as the direction of projection at the first stage. Y -axis is the error rate on the test data. The X-axis in the top-left panel is the number of features selected in the models. The X-axes in the top-right and bottom-left panels are the indices of the tuning parameter; each index i (i∈{0,…,99}) corresponds to λ_max−i(λ_max−λ_min)/99; λ_max is the maximum value of the tuning parameter λ calculated by default by R-package glmnet. λ_min=λ_maxλ_min.ratio and λ_min.ratio=0.01, which is the default value of glmnet. The axis in the bottom-right panel is the cost that corresponds to the parameter c in the R-package e1071.