Clustering of Social Media Messages for Humanitarian Aid Response during Crisis
This work provides incremental improvements for crisis response teams by enhancing automated message filtering from social media.
The authors tackled the problem of classifying social media messages for humanitarian aid during crises, showing that recent deep learning and NLP methods outperform prior approaches for informativeness classification and are also effective for two related sub-tasks.
Social media has quickly grown into an essential tool for people to communicate and express their needs during crisis events. Prior work in analyzing social media data for crisis management has focused primarily on automatically identifying actionable (or, informative) crisis-related messages. In this work, we show that recent advances in Deep Learning and Natural Language Processing outperform prior approaches for the task of classifying informativeness and encourage the field to adopt them for their research or even deployment. We also extend these methods to two sub-tasks of informativeness and find that the Deep Learning methods are effective here as well.