CS229 Class Project
To dos before Wednesday Week 6:
- PRAW
- Create notebook
- Shows attributes we have access to
- Sizes of datasets
- What limitations do we have on data access etc.
- NLTK
- Create notebook
- What can we do with the sentiment analysis API?
- Literature review
- Read previous papers to have a sense of what has already been done
- Create Overleaf document
References:
- A few preliminary resources:
- Sentiment Analysis on Reddit News Headlines with Python’s Natural Language Toolkit (NLTK)
- The New American Lexicon (book page 165; this is the document advising the Republican party on language change)
- A few basic papers on sentiment analysis for climate change language:
- Climate Change Sentiment on Twitter: An Unsolicited Public Opinion Poll (highest citation count paper on the subject that I could find)
- Tracking Climate Change Opinions from Twitter Data
- Climate Sentiment Analysis on News Data (R notebook) (perhaps not the most useful analysis, esp because in R, but perhaps interesting to look at their datasets)
later: 7a) clearly define the machine learning question: could be building a sentiment analyzer based off this data; take off-shelf sentiment analyzer, train classifier to predict overall sentiment would be in response to sentence within community.