Exercise $1$: sentiment recognition

Attached Files:

File nyt$\_200811.$txt $(2\mathrm{.}911$ MB$)$

Go to the attached document. Come up with $1$ sentence that the classifier gets wrong. Explain, the best you can, what is causing the error?

$1.$ load the text from the attached document.

$2.$ use UNIX$/$Linux scripting to find, based on most commonly used words, what is the text about. Justify your answer.

$3.$ useful commands: tr$,$ sort, uniq, e$.$g$.$ see what happens with $\text{'}$tr $-$sc $\text{'}$A$-$Za$-$z$\text{'}\text{'}$

$\text{'}<$ nyt$\_200811.$txt $|$ tr $\text{'}[$:upper:$]\text{'}\text{'}[$:lower:$]\text{'}|$ sort $|$ uniq $-$c$\text{'}$