Real-time assessment of human loss in disasters based on social media mining and the truth discovery algorithm

Abstract

The cycle of traditional disaster loss assessment methods is too long because the assessment results need to be passed among multiple issues. Such a long cycle causes relevant departments to be unable to make emergency decisions in a timely manner. This paper builds a dynamic and real-time model for disaster loss assessment based on social media data. This model presents a method for mining quantitative human loss information from social media data and a truth discovery algorithm for handling social media data conflicts. An experiment using the actual data of the Jiuzhaigou earthquake is conducted to verify the model. The results confirm that the method proposed in this study can accelerate the process of reporting traditional human loss information between multiple levels and departments and assess human loss in a real-time and dynamic manner. The findings are important for providing timely assistance to official agencies and the public.