A human-centric approach to classify post-disaster recovery priorities

Giant natural, extreme events inflict all aspects of people’s lives. These immense catastrophes get the lives of people, disrupt social interactions, ruin infrastructural components, and destroy residential properties, as well as several other short or long-term social, economic, and geographical effects. Though, the main task of policymakers is to return the lives of people back to normal condition after being hit by disasters, this requires accurate assessments of peoples priorities. Each individual of the community perceives damages of the disaster differently based on their personal characteristics such as age, income, ownership, and many other factors named as internal attributes. Adding up the complex consequences of natural extreme events to the broadness of internal attributes, detecting of survivors priorities becomes a highly complex problem. Fortunately, day-to-day growth of social media applications with an incredible penetration in all aspects of life provides a unique substrate for researchers to detect thoughts, beliefs, and opinions of disaster survivors. On the other hand, due to the broadness of social media users alongside with complexities of disaster impacts, valuable social media data was not broadly utilized in post-disaster studies. As such, in this study, a step-by-step journey implemented to picture a thematic shape of communities, their priorities, and decision-making with the underlying contribution of internal attributes. Literature review of authentic studies, qualitative and quantitative methods, as well as Machine Learning Algorithms, were performed to reach this goal.