An Attempt at Quantifying Disaster Damage Based on the Use of Collective Intelligence

抄録

It is extremely important to evaluate the extent of socioeconomic damage to draw up a disaster preparation program and determine the specifics and their scales when implementing disaster prevention and reduction measures before the occurrence of disasters. Yet, it has only been possible in the past to represent the entire damage costs by part of the damage that could be quantitatively evaluated. This necessarily resulted in underestimation. In this study, the author developed a method in which the descriptions of damage scenarios are collected from over a thousand people and the damage costs evaluated by examining the frequency of used words. Specifically, a questionnaire survey was first conducted to extract the people’s views of disaster, and the frequencies of specific words used in these descriptions of disaster were used as indicators of damage. It was then assumed that the difference in the appearance of these words in newspaper articles before and after an actual disaster represents the damage impact, and this difference was used to estimate the damage costs. The results suggested the possibility of developing an evaluation method based on collective intelligence, as well as the need to improve and refine the method in the future.