predbayescor: Classification rule based on Bayesian naive Bayes models with feature selection bias corrected

This software is used to predict the binary response based on high dimensional features, for example gene expression data. The data are modelled with Bayesian naive Bayes models. When a large number of features are available, one may like to select only a subset of features to use, typically those features strongly correlated with the response in training cases. Such a feature selection procedure is however invalid since the relationship between the response and the features will appear stronger. This package provides a way to avoid this bias and yields well-calibrated prediction for the test cases.

Downloads: