Landslide Tracking Using Social Media

Earthquake-induced landslide in Japan in April 2016
The site of a large, earthquake-induced landslide is seen in Minamiaso, Kumamoto Prefecture, on April 16, 2016 (Mainichi)

Some clever researchers have devised an algorithm to collect information on landslides occurring around the world by analyzing social media posts. They use a variety of search terms to flag a post as being potentially related to landslides, then they use artificial intelligence to analyze imagery to determine if it is a landslide event and to scrape relevant information, including location data. The Landslide Blog has more on this and a link to the paper by Catherine Pennington from the BGS and a team of international collaborators.

Be the first to comment

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.