What can geocoded tweets tell us about the rapid spread and impact of Covid-19 in South Asia?
New from @obtusatum bellingcat.com/resources/2021…
New from @obtusatum bellingcat.com/resources/2021…
@obtusatum By combining the results of many geocoded queries, Bellingcat was able to make an approximate map of the geographic distribution of tweets looking for urgent assistance across India and its South Asian neighbours, many of which are experiencing their own coronavirus crises
@obtusatum India in particular has experienced a dramatic recent rise in Covid-19 cases, as we explored previously bellingcat.com/news/2021/04/2…
@obtusatum Pulling together this Twitter data helps visualise the scale of this issue and which areas seem to have been most impacted.
It also reveals some new subtleties of how Twitter provides location data, and how that data can be used by researchers.
It also reveals some new subtleties of how Twitter provides location data, and how that data can be used by researchers.
@obtusatum While the techniques described within this story come with caveats, they can also be adapted and used in other projects.
Documentation of Twitter’s treatment of location data could help researchers interpreting Twitter data in many contexts.
Documentation of Twitter’s treatment of location data could help researchers interpreting Twitter data in many contexts.
@obtusatum Access to Twitter’s API, while useful, is also not necessary given free tools such as TWINT can be used to download thousands and even millions of tweets relatively quickly 
@obtusatum For full details on how to do so, read the full article here bellingcat.com/resources/2021…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
