Post-disaster recovery assessment using sentiment analysis of english-language tweets: A tenth-anniversary case study of the 2010 Haiti earthquake

This study uses social media (SM) as a resource to track the progress of the post-disaster recovery of the 2010 Haiti earthquake. Text data from tweets on SM undergo sentiment analysis, providing a promising tool for understanding earthquake recovery. Sentiment analysis allows for the processing of unstructured data, classifying people's opinions into positive, negative, or neutral polarity. Therefore, the use of social media (SM) has enhanced earthquake reconnaissance teams' capacity to collect data beyond the emergency phase, allowing for this study's assessment of recovery progress over a decade.

From data collected around the tenth anniversary of the earthquake, and using human expertise to fine-tune automatic classification methods, the study concluded that the anniversary date is the best time to collect data related to this event. In the sample, 56.3% of the tweets were classified as negative, followed by positive (27.3%), neutral (8.2%), and unrelated (8.1%). In this research, the study concludes that the assessment of the recovery progress based on data collected from Twitter is negative. The negative polarity of tweets reports the role of the Clinton Foundation, the construction of the industrial park in Caracol, the food shortages, the cholera epidemic, child trafficking, hurricane Mathew, and the low number of houses constructed, the poor execution of financial resources raised, the decamped of international development programs and the political chaos.