Support

Support Options

Submit a Support Ticket

 

Model Validation Document for "A Meta-Analysis of Carbon Nanotube Pulmonary Toxicity Studies – How Physical Dimensions and Impurities Affect the Toxicity of Carbon Nanotubes"

By Jeremy M Gernand1, Elizabeth Casman1

1. Carnegie Mellon University

Download (PDF)

Licensed according to this deed.

See also

No results found.

Published on

Abstract

This document contains model learning statistics, and structure of the models utilized in the paper "A meta-analysis of carbon nanotube pulmonary toxicity studies – How physical dimensions and impurities affect the toxicity of carbon nanotubes." This information is meant to supplement and support the explanations and conclusions reached in that paper.

This document includes the detailed structure of the pruned regression tree models (Figures 1 through 3) as well as the tree's error performance as a function of model growth (Figures 4 through 7). The random forest model performance versus model growth for each of the 4 output measures is also included (Figures 8 through 11).

The stepwise random forest models and their performance as a function of included variables are displayed in Figures 12 through 15. The random forest generated dose-response profiles and the effects of cobalt impurities are shown in Section 6 (Figures 16 through 18).

The MATLABTM code that creates these regression tree and random forest model objects is included in the final section. The data used to train the models can be found at https://nanohub.org/resources/13515.

Cite this work

Researchers should cite this work as follows:

  • Jeremy M Gernand; Elizabeth Casman (2012), "Model Validation Document for "A Meta-Analysis of Carbon Nanotube Pulmonary Toxicity Studies – How Physical Dimensions and Impurities Affect the Toxicity of Carbon Nanotubes" ," http://nanohub.org/resources/15901.

    BibTex | EndNote

Tags

nanoHUB.org, a resource for nanoscience and nanotechnology, is supported by the National Science Foundation and other funding agencies. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.