Twitter Sentiment Analysis with Text Summaries and Keywords

This Python code analyzes thousands of tweets using 2 sentiment analysis libraries (TextBlob and VADER), summarizes each classification of tweets using 4 text summary tools (LexRank, Luhn, and 2 versions of LSA), and now lists the stopwords-scrubbed keywords that accompany the given search term.

Each classification (negative, positive, neutral, unknown) of tweets now has its own keywords list. The results, once again, are both displayed on-screen and saved in a text file.

The next update will likely include:

• code cleanup; I’ve been focusing on getting it to work, so I have some sections that I can hopefully shorten with new functions
• more stopwords; I’m not satisfied with the keywords results, so I want to try to add words to the set that they are scrubbed against
• label consistency; I’ve got “LexRank Negative” and “Negative Keywords,” so I want to review all the labels for consistency
• comments; I have comments throughout the code, but I want to add more and make sure they are sufficiently explanatory

I am otherwise out of ideas as to how to further enhance this particular code. I’m sure my Python could be better, but I am unaware of any other features that would be worth adding. I would like to find other APIs to apply it to, but that’s about it.