Question $3($Mandatory$)(30$ points$)$

Open notes, open WWW$,$ work on your own. You should use R and load the

following packages: tm$,$ stringr, and wordcloud.

$($Place final answers in provided cells adjacent to questions.$)$

Use R to mine the book Peoples' History of the U$.$S$.$A$.1492$ to Present by Howard

Zinn.

The text is stored in the file $\text{'}$ UShistory.txt$.$ Answer the questions below.

a$)$ How many words are in the book before cleaning up the text? $($hint: use the

str$\_$split$()$ and unlist$()$ routine you performed in the lab$)$

$($round to nearest $10,000,$ for example $346,344$ would be $350,000)$

A

b$)$ Clean the text in this exact order. Remember to use the unlisted object you

created from above to start this process:

remove punctuation$/$special characters

remove digits

make everything lowercase

remove stopwords using the 'english' stopwords in package tm

remove extra whitespace

Now, what is the most frequent term used in this book?

A

c$)$ Using your cleaned text document from question $b$ along with the Positive and

Negative word lexicons provided on blackboard $($or already saved in your Google

Colab$),$ compute the sentiment score for this book. That is$,$ what is the result of the

sum$()$ function for negative matching subtracted from the sum$()$ function for positive

matching? $($round to nearest $100,$ for example $8,342$ would be $8,300)$