Development of Heavy Rain Damage Prediction Model Using Machine Learning Based on Big Data

Self Cite

In this study, we developed heavy rain damage prediction functions using three machine learning techniques (support vector machine, decision tree, and random forest) for the Seoul Capital Area, South Korea. Data on damage caused by heavy rain were used as the dependent variable for the development of the heavy rain damage prediction function, and weather observation data were used as the independent variables. When we compared the results, the best function was the support vector machines based on weather observation data of the past two days. Compared to the linear regression model used primarily in previous studies, the results showed that the functions using machine learning techniques were mostly predictable. Therefore, it was judged that the machine learning techniques could be applied to disaster management areas. Also, it is believed that using the heavy rain damage prediction function developed in this study can help reduce damage through proper disaster management before the damage occurs.

Section: 서 론unclassified

Section: 머신러닝 기법unclassified

Section: 의사결정나무unclassified

See 1 more Smart Citation

Development of Heavy Rain Damage Prediction Functions in the Seoul Capital Area Using Machine Learning Techniques

Choi¹,

Kim²,

Kim³

et al. 2018

Self Cite

“…1. 서 론 전 세계적으로 지구온난화로 인해 자연재난의 발생 빈도 및 규모가 증가하고 있으며, 재난으로 인한 심각한 경제적 손실과 인명피해가 발생하고 있다 (UNISDR, 2016;Choi et al, 2018). 특히, 경제가 발전함에 따라 재난으로 인해 직접적인 피해를 받은 산업이 다른 산업에 미치는 간접 피해도 점점 커지고 있으며, 국가 경제에 막대한 피해를 발생시켜서 경제성장을 저해하는 요인으로 나타나고 있다 (Ryu et al, 2012;Choi et al, 2017a).…”

unclassified

Impact Assessment of Natural Disaster Damage on the National Economy Using the Input-Output Model

Choi¹,

Kim²,

Lee³

et al. 2019

Self Cite

In this study, the input-output model (IO model) was used to quantitatively estimate indirect damages caused to industries by those that are directly impacted by natural disasters; the corresponding impact on the national economy was also evaluated. Indirect damages from other industries due to direct damages of natural disasters were mainly observed in the manufacturing industry, while the impact on the service industry was minimal. For example, natural disasters have led to the largest direct damages to the construction industry, as well as to industries involved in the manufacture of textiles, leather, and transport equipment. The results of the evaluation of the impact on the national economy, by comparing the average annual GDP and natural disaster damage, showed that the total damage (direct damage + indirect damage) is almost three times higher than when only considering direct damage. Therefore, in order to implement effective disaster management, it is necessary to consider indirect damages by industries along with direct damages during decision making.

“…자연재해로 인한 피해예측방법에는 손실⋅손상함수를 이용하여 피해를 예측하는 방법 (Kim et al, 2014;과 수문기상학적 인자와 피해액 사이의 상관관계를 이용하여 산정하는 방법 (Kim et al, 2014;Kwon and Chung, 2017;Choi et al, 2018;Choo et al, 2018;Y. Kim et al, 2018)이 있다.…”

unclassified

Development of Typhoon Damage Prediction Function Using a Logistic Distribution

Kim¹,

Choi²,

Kim³

2019

Self Cite

The purpose of this study is to develop a typhoon damage prediction function considering the correlation between damage amount due to typhoons and hydrometeorological factors estimated from disaster statistics in coastal districts. As a typhoon approaches, heavy rain, gale, and storms jointly and simultaneously cause severe damage. A total of 32 variables were selected to derive the function, such as factors that affect rainfall-runoff, maximum wind velocity, and central typhoon pressure. As observed, typhoon damages have a wide range-from hundreds of thousands of KRW to more than KRW 1 trillion-and accordingly have been classified into large and small groups depending on the amount of damage; thereafter, the typhoon damage prediction function was developed for each group. The logistic regression analysis was used for group classification of typhoon damages. Furthermore, the principal component regression analysis was employed for the typhoon damage prediction function. The principal component analysis effectively solved the problem of multicollinearity due to high correlation between 32-dimensional independent variables and rainfall-runoff related variables. According to the results of the typhoon damage prediction function, the Normalized Root Mean Squared Error (NRMSE) of the large group was 12.4%, while that of the small group was 18.5%.