Social media and crowdsourcing platforms have become a valuable tool for quickly collecting large amounts of disaster data. Image and text data posted on Twitter, Instagram or in crowdsourcing platforms such as LastQuake app can support emergency response operations, earthquake reconnaissance missions and the assessment of recovery processes after earthquakes in the long term. However, it is necessary to use NLP techniques such as sentiment and topic analysis to extract meaningful information from the unstructured text data from social media. Currently text data from the 2019 Albania, 2020 Zagreb, 2020 Aegean earthquake and the 10th anniversary of the earthquakes in L’Aquila, Chile, Haiti and Nepal have been classified using supervised and unsupervised methods to assess the emergency response and the recovery after earthquakes.
Invited Speaker: Diana Contreras Mojica