Classification in the form of dendrograms can be done with following commands (only first 10 rows of dataset are used here to simplify the plot):
> fit = hclust(dist(bwdf[1:10,]))
It shows that rows named '0-1' an '0-2' are very similar to each other.
Following command which uses varclus (variable clustering) function of Hmisc package can be used to create dendrogram of different variables:
It can be seen that low and bwt are related, as are age and ftv and also race and smoke.
mass package: Venables, W. N. & Ripley, B. D. (2002) Modern Applied Statistics with S. Fourth Edition. Springer, New York. ISBN 0-387-95457-0
Frank E Harrell Jr, with contributions from Charles Dupont and many others. (2015). Hmisc: Harrell Miscellaneous. R package version 3.16-0. http://CRAN.R-project.org/package=Hmisc