Using neural networks for classification
nnet package contains the
nnet function for classification using neural networks.
If you have not already installed the
caret packages, install them now. Download the data files for this chapter from the book's website and place the
banknote-authentication.csv file in your R working directory. We will use
class as our target or outcome variable, and all the remaining variables as predictors. Using Neural Networks requires all the independent/predictor variables to be numeric and the dependent variable or outcomes to be 0-1. However, the
nnet function does all the work of generating dummies (contrasts) and correctly handles categorical outcome variables.
How to do it...
To use Neural ...