Abstract
This work presents different data visualization approaches to recover from a natural disaster using social media data. During natural disasters, it is essential to analyze the situation and react quickly to mitigate loss and chaos. It is essential to recover from the disaster as quickly as possible before things run out of control. Usually, management or officials who work in this field are often going clueless, considering the chaos they need to handle. We need to come up with accurate evidential analysis to help them distribute resources at the right time and right place. Data visualization is considered one of the quickly emerging and powerful ways of handling big data and real-world problems. I am considering the IEEE VAST 2019 Challenge [1] dataset, the fictional social media data, to solve this problem efficiently with the help of different data visualization techniques. Nowadays, most of the people are active in social media, and with the minimal preprocessing of such data, we can determine what exactly is going on with people’s life. So, it is essential to make use of such social media data during disaster recovery. I have developed an interactive visualization tool which could help us analyze the situation of the place considering different timestamp. This tool can give an insight into the different issues people could face at a different location and help the disaster management team to allocate different resources as per the needs and preferences of the people in real quick time thus mitigating the loss. I have used D3.js, JavaScript, Python for interactive visualization of social media data, and data preprocessing. Also, I have analyzed different scenarios to understand how the disaster affected the place and how effectively we can allocate essential resources with the help of static visualizations using Tableau software. Thus, with the blend of powerful data visualization and social media data, I am displaying different practical approaches to mitigate the loss and make a significant contribution to the field of disaster recovery.