Your assignment is to apply a text analysis method to a corpus of your choice. You may use the ATS corpus, the WPA former slave narratives, or the Tracts for the Times corpora provided, or you may find your own corpus. Note that it is not necessary for the purposes of this assignment to use a very large corpus.
You should start by applying a term-frequency and TF-IDF analysis to your corpus (see the quanteda package in particular). Then you should also analyze the corpus using either topic modeling, named entity recognition (the cleanNLP package), sentiment analysis, word vectors, or text reuse. The quanteda package also contains a number of sample analyses.
For this assignment, you can turn in an analysis notebook that explains your process and what you have found.
install.Rwill help you install the key packagesats/contains 20 sample tracts printed by the American Tract Society- The files named
*-example.Rshow examples of doing different kinds of analysis.