Post Disaster Situation Awareness and Decision Support Through Interactive Crowdsourcing

Abstract In today's world, online social media has established itself as one of the most important information carriers, and that applies in the field of disaster management as well. However, posts in online social media like Twitter and Facebook are mostly unstructured, and it is extremely difficult to extract meaningful information from such unstructured posts. Therefore, disaster management authorities face problems in using these posts for proper situation analysis and decision-making. To alleviate this problem, we propose a decision support framework that collects situational information through interactive crowd-sourcing using SMS from the “crowd” present at the disaster site, and summarizes such responses to have situational awareness and appropriate decision-making regarding damage or need assessment. We evaluated our proposed system using 3000 real-time interactive responses from a disaster-prone coastal area of India (named, the Sunderbans). Since the use of smart-phone in rural India is significantly low, SMS from the “crowd” were used to form the core information repository for post-disaster situation analysis.