Humanitarian Assistance Ontology for Emergency Disaster Response

Massive amounts of open data related to a crisis aren't fully used to identify humanitarian needs because most of the data exists in an unstructured format, thus hindering machines in its interpretation. Automatic processing of crisis data in a short time period would provide useful information to decision makers. To address these problems, this article presents a method for merging ontologies and logic rules to represent humanitarian needs and recommend appropriate humanitarian responses. The main advantage of the method is that it identifies humanitarian needs and prioritizes humanitarian responses automatically, including health survival issues and possible responses by creating a working system during the evolution of the crisis. The method is implemented on 125 real data reports from the Hurricane Wilma crisis and compared to the actual support provided.