I'll be placing scripts and script fragments here that I've used for analyzing Tweets that relate to sake (beverage). Whatever I put up here will be the analysis scripts, not the data collection. A non-exhaustive list of options for data collection are below. Choose what best fits your problem.
- Sign up for a Twitter developer account and use their API (easy but limits the timeframe of searches to recent history)
- Twitterscraper (well built Python package that has command-line-interface, easy to use and not limited in timeframe)
- Pay a company for historical Twitter data (
$—$ $$)
- List of user locations that are uninformative or in some way give bad geocoding. Includes associated R code for removal.