Machine Learning for Damage Classification, Risk Mitigation and Post-earthquake Management

Date

2024-01-01

Author

Di Michele, F.
Giannopoulou, O.
Stagnini, E.
Pera, D.
Rubino, B.
Aloisio, R.
Askan Gündoğan, Ayşegül
Marcati, P.

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

0
views

0
downloads

In recent years, we have witnessed a steady increase in the amount of data collected and made available to the scientific community. Simultaneously, Artificial Intelligence (AI) shows great potential in transforming data to knowledge offering increasingly more accurate tools to analyze, interpret and extract information from the data. In this context, data-driven approaches, in the fields of seismology, geophysics, and earthquake engineering show enormous promise. In this work a Machine Learning (ML) study is presented, based on a dataset containing around 3000 buildings damaged by the 2009 L’Aquila earthquake. This event was the first in a series of strong earthquakes that hit central Italy, resulting in many casualties, and having enormous economic and social impact. Each building in the dataset is described by 22 characteristics. Among them the damage level, divided, into six classes, from D0, corresponding to no damage, to D5, corresponding to heavy damage or collapse. We employ a Random Forest based algorithm to predict the level of damage accounting for different combinations of damage levels. As a case study we consider the following binary target variables-D0-D1-D2-D3 (no to medium damage) and D4-D5 (serious to heavy damage)-D0-D1 (no to light damage) and D2-D3-D4-D5 (moderate to heavy damage)-D0-D1-D02 (light to moderate damage) and D3-D4-D5 (medium to heavy damage).

Subject Keywords

L’Aquila 2009 earthquake, Machine learning, Risk mitigation, Seismic damage prediction

URI

https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85197470976&origin=inward
https://hdl.handle.net/11511/110245

DOI

https://doi.org/10.1007/978-3-031-57357-6_16

Conference Name

7th International Conference on Earthquake Engineering and Seismology, ICEES 2023

Collections

Department of Civil Engineering, Conference / Seminar

Citation Formats

F. Di Michele et al., “Machine Learning for Damage Classification, Risk Mitigation and Post-earthquake Management,” Antalya, Türkiye, 2024, vol. 401 LNCE, Accessed: 00, 2024. [Online]. Available: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85197470976&origin=inward.