Weakly-supervised Fine-grained Event Recognition on Social Media Texts for Disaster Management
This addresses the time-critical need for disaster management by enabling rapid, high-quality event classification from noisy social media texts.
The paper tackles the problem of rapidly building classifiers for fine-grained event recognition on social media during disasters, achieving results where weakly supervised classifiers trained with only 1-2 person-hours of human supervision outperform supervised classifiers trained with over 50 person-hours and 10,000 annotated tweets.
People increasingly use social media to report emergencies, seek help or share information during disasters, which makes social networks an important tool for disaster management. To meet these time-critical needs, we present a weakly supervised approach for rapidly building high-quality classifiers that label each individual Twitter message with fine-grained event categories. Most importantly, we propose a novel method to create high-quality labeled data in a timely manner that automatically clusters tweets containing an event keyword and asks a domain expert to disambiguate event word senses and label clusters quickly. In addition, to process extremely noisy and often rather short user-generated messages, we enrich tweet representations using preceding context tweets and reply tweets in building event recognition classifiers. The evaluation on two hurricanes, Harvey and Florence, shows that using only 1-2 person-hours of human supervision, the rapidly trained weakly supervised classifiers outperform supervised classifiers trained using more than ten thousand annotated tweets created in over 50 person-hours.