MOVIEDATA.

CSV

using Rstudio

$2.$ Do you want to remove any additional user$-$defined stop words? If so$,$ please create the stopword$1$ list, list those words here, and remove them from your DFM$.(5$ points$)$

$3.$ A movie critics further examines the dataset, and recommend removal of the following commonly used terms: $(5$ points$)$

film, movi, play, even, just, go$,$ get, like, time, make, charact, scene, show, $1,2,$ year, come, may, john

Please create the stopword$2$ list and use dfm$\_$remove to further remove those common terms from your DFM$.$

Please create a word cloud with the $150$ most used terms. $(8$ points$)$

Paste the visualization below.

Please summarize a few common things covered by the movie synopses.

Please further remove highly infrequent terms. $(8$ points$)$

Specifically, we only keep terms that appear in the entire corpus at least $15$ times and appear in at least $5$ different movies.

What is the dimensionality of the trimmed DFM$?$

Which movie is most similar to Movie $91("$Batman & Robin"$),$ based on the correlation similarity measure? $(8$ points$)$

What are the top $8$ terms that are most related to "school", based on the cosine similarity measure. $(8)$ points$)$

Please perform topic modeling with $4$ topics and keep $8$ most relevant terms per topic. $(9$ points$)$

Provide the term$/$beta plots for four topics.

Try your best to summarize those four topics.