Humanitarian assistance ontology for emergency disaster response

Abstract

Humanitarian Assistance Ontology is an ontology that covers crisis concepts and its relations based on humanitarian standard needs and response. HAO consist of two sub-ontologies, Crisis Identification & Crisis Response sub-ontology, Crisis Identification Ontology to determine the humanitarian needs from the input data and Crisis Response Ontology to determine the response recommendations for the actual humanitarian needs. The HAO system consists of preprocessing stage, and inferring stage. Preprocessing stage includes Named Entity recognition to identify location entity, context recognition process using C/NC value & context recognition algorithm to elicit concept entities and its contexts and context mapping to map extracted entities into crisis identification sub-ontology. A set of logic rules is used to unify these two ontologies and to generate the humanitarian response recommendations. The method is implemented on real data from the Hurricane Wilma crisis. The result in the hurricane Wilma crisis shows the potential abilities to identify the humanitarian needs in specific places and to infer humanitarian responses in real time.