Tweets with emergency requests during natural disaster are rare but very valuable to classify.
•
Machine learning models trained on Hurricane Harvey data are used to detect urgent tweets.
•
CNN, SVM, and multilayer perceptron trained on word embeddings achieve F1 over 0.86
•
Average word embeddings are a novel and effective feature type for non-neural disaster models.
•
Classifiers could be effectively used by first-responders to identify those needing rescue.
Abstract
Social media is increasingly used by people during large-scale natural disasters to request emergency help. Previous work has had success in applying machine-learning classifiers to detect tweets in coarse-grained categories, such as disaster type and relevance. However, there is a dearth of work that focuses on detecting tweets containing requests for help that are actionable by first responders. Using over 5 million tweets posted during 2017's Hurricane Harvey in Houston, U.S., we show that though such requests are uncommon, their often life-or-death nature justifies the development of tweet classifiers to detect them. We find that the best-performing classifiers are a convolutional neural network (CNN) trained on word embeddings, support vector machine (SVM) trained on average word embeddings, and multilayer perceptron (MLP) trained on a combination of unigrams and part-of-speech (POS) tags. These models achieve F1 scores of over 0.86, confirming their efficacy in detecting urgent tweets. We highlight the utility of average word embeddings for training non-neural models, and that such features produce results competitive with more traditional n-gram and POS features.
