ONTOLOGY-BASED TECHNOLOGIES FOR DISASTER PREPAREDNESS, RESPONSE AND RECOVERY Panagiota Masa 1 , Georgios Tzanetis 2 , Spyridon Kintzios 3 , Georgios Meditskos 4,5 , Stefanos Vrochidis 6 , Iosif Vourvachis 7 , Lorenzo Nerantzis 8 , Ioannis Kompatsiaris 9 1,2,3,4,6,9 Information Technologies Institute, Centre for Research and Technology Hellas, (Greece) 5 School of Informatics, Aristotle University of Thessaloniki, Greece 7,8 Hellenic Rescue Team (E-mail: gmasa@iti.gr, tzangeor@iti.gr, sp.kintzios@iti.gr, gmeditsk@csd.auth.gr, stefanos@iti.gr, i.vourvachis@hrt.org.gr, l.nerantzis@hrt.org.gr, ikom@iti.gr) ABSTRACT Natural disasters such as forest fires, earthquakes, floods and heat waves have a tremendous impact on the economy, the environment and the people. Every year, manmade and natural disasters lead to human and property losses and resettlement, the disintegration of infrastructures and degradation of society's resilience. The key role in the efficient management of such a crisis is information and knowledge management. The fusion of heterogeneous information, sensors and data processing is the stepping stone for every system architecture. This paper proposes the use of ontology-based technologies for disaster preparedness, response and recovery to ensure effectiveness. Innovative semantic structures that have been used in relevant EU projects are explored and adapted in the respective framework. Keywords: Semantic Web, Ontology, Semantic Reasoning, Decision Support, First responders. 1. INTRODUCTION Natural and manmade disasters are becoming more frequent and more devastating. The use of technology could leverage the response and make societies more resilient. Data acquisition, taxonomy and fusion from multimodal services (video and image analysis, Earth Observation-EO tools, social media, weather forecasts, wearables, etc.) can provide an advantage for crisis management. Decisions could be supported by facts and analysis (even trial and error) instead of being intuitive. Human fatigue and limited cognitive could be overpassed by Disaster Management Systems (DMS) and Disaster Support Systems (DSS). Early warning messages provide alerts to prepare and protect citizens and first responders. Meanwhile, wearable equipment monitors first responders’ status and reflects the feasibility to fulfill their missions. A promising approach to building complex systems is the use of Semantic Web technologies [1]. These technologies, with their ability to integrate heterogeneous data from various sources, offer a significant amount of potential to extend the information available on the Web by giving the information an understandable meaning that can be used by applications for assisting emergency management. Information is represented via ontologies while semantic reasoning schemes run on top of the ontologies and facilitate situational awareness and analysis for decision support. The use of semantic technologies for disaster management has been used in different disaster cases, e.g. earthquakes [2], in order to reduce the response time in disaster detection scenarios, and wildfires [3], in order to enhance knowledge dissemination and operational stakeholder preparedness. The inference process is essential wherever there is a need to identify potential data inconsistencies, evaluate data, identify new relationships, and build new knowledge.