Spatio-temporal mining of keywords for cross-social crawling of emergency events
Andrea Autelitano, Barbara Pernici, Gabriele Scalia · Virtual Community of Pathological Anatomy (University of Castilla La Mancha) · 2018
Being able to automatically extract as much relevant posts as possible from social media in a timely manner is key in many activities, for example to provide useful information for rapidly creating crisis maps during emergency events.While most of the social media support keyword-based queries, the amount and the accuracy of the retrieved posts depends largely on the keywords employed.The goal of the proposed methodology is to automatically and dynamically extract relevant keywords for ongoing events in order to ultimately crawl as much relevant posts as possible.This is accomplished taking into account the spatio-temporal features of the monitored event to better characterize it during its evolution and through cross-social crawling in order to exploit the specificities of a social media on the others.The methodology has been implemented on Flickr and YouTube and evaluated on two recent major emergency events demonstrating a large increment in the number of crawled posts with respect to using simple generic keywords and their high relevance for the scope.