Analysis of the Seismic Destructive Force and Building Features with Tree-Based Machine Learning

Yingquan Lei

2022

Abstract

Earthquakes have long been highly destructive and can cause huge economic losses and casualties. It is natural that people hope to predict earthquakes in advance so as to avoid losses. Whereas earthquake is a very complex geographical phenomenon to predict and the data required is not sufficient currently, so, unfortunately, there is no accurate prediction method yet. However, there is something worth exploring from the perspective of building characteristics, which is more controllable and easier to study compared to the mysterious earthquake. And so far, little research has been conducted about the connection between building features and earthquake damage. In this paper, first, with the help of some python visualization tools, several representative building features are analyzed, giving possible solutions to targeted disaster relief as well as how to improve the seismic ability. The first few most important features are found and the significance is quantified. Then machine learning is involved to predict the damage, and the optimal algorithm has a 72% accuracy rate.

Paper Citation

in Harvard Style

Lei Y. (2022). Analysis of the Seismic Destructive Force and Building Features with Tree-Based Machine Learning. In Proceedings of the 2nd International Conference on Public Management and Big Data Analysis - Volume 1: PMBDA; ISBN 978-989-758-658-3, SciTePress, pages 134-141. DOI: 10.5220/0012071100003624

in Bibtex Style

@conference{pmbda22,
author={Yingquan Lei},
title={Analysis of the Seismic Destructive Force and Building Features with Tree-Based Machine Learning},
booktitle={Proceedings of the 2nd International Conference on Public Management and Big Data Analysis - Volume 1: PMBDA},
year={2022},
pages={134-141},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012071100003624},
isbn={978-989-758-658-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Public Management and Big Data Analysis - Volume 1: PMBDA
TI - Analysis of the Seismic Destructive Force and Building Features with Tree-Based Machine Learning
SN - 978-989-758-658-3
AU - Lei Y.
PY - 2022
SP - 134
EP - 141
DO - 10.5220/0012071100003624
PB - SciTePress