still proposal submission

[supertrashpanda]

Project Proposal.pdf

[JonathanReeve]

Some interesting ideas here. I'd suggest narrowing your focus a little, because movies-in-general is so incredibly broad of a topic. Choosing just one or two of your data sets and/or analysis ideas would be a good place to start.

There are some movie data sets out there which align both IMDB and RottenTomatoes ratings, which I think is interesting. For instance, that would allow you to ask questions like:

• What movies are loved by critics (RottenTomatoes), but hated by average users of IMDB?
• What movies are loved by IMDB users but hated by critics?

You could also look at things like: what movies don't "translate well," that is, what movies are loved in one country and disliked in another? You might be able to find this by looking at information about the reviewers in IMDB.

Personally, I've always been interested in movie titles which are hard to translate: which movies are known by lots of other names, in lots of other languages? What movie titles are translated very differently?

If you're going to analyze movie dialogue, that's a good angle, and I'd suggest focusing entirely on that, if you choose to. That would open up lots of interesting questions, like:

• What frequent words are correlated with high ratings? Are there any?
• What is the average number of speaking roles (people who speak) in movies? How much does that change, and is it correlated with rating?
• What is the average length of utterances (lines of dialogue) in movies? How does that correlate with the movie's genre? Does it correlate with ratings?
• What are the words that occur most in dialogue, sorted by MPAA rating? That is, what kinds of words are used in G-rated movies, and how are those different from R-rated movies? (This would make a good side-by-side set of word-clouds.)