Comparison of machine learning tools for damage classification: the case of L’Aquila 2009 earthquake

Download

index.pdf

Date

2023-1-01

Author

Di Michele, F.
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

187
views

66
downloads

On April 6, 2009, a strong earthquake (6.1 Mw) struck the city of L’Aquila, which was severely damaged as well as many neighboring towns. After this event, a digital model of the region affected by the earthquake was built and a large amount of data was collected and made available. This allowed us to obtain a very detailed dataset that accurately describes a typical historic city in central Italy. Building on this work, we propose a study that employs machine learning (ML) tools to predict damage to buildings after the 2009 earthquake. The used dataset, in its original form, contains 21 features, in addition to the target variable which is the level of damage. We are able to differentiate between light, moderate and heavy damage with an accuracy of 59%, by using the Random Forest (RF) algorithm. The level of accuracy remains almost stable using only the 12 features selected by the Boruta algorithm. In both cases, the RF tool showed an excellent ability to distinguish between moderate-heavy and light damage: around the 3% of the buildings classified as seriously damaged were labeled by the algorithm as minor damage.

Subject Keywords

Seismic damage prediction, Machine learning, L'Aquila 2009 earthquake, Ground motion, FEATURE-SELECTION, PREDICTION, Ground motion, L’Aquila 2009 earthquake, Machine learning, Seismic damage prediction

URI

https://hdl.handle.net/11511/102507

Journal

Natural Hazards

DOI

https://doi.org/10.1007/s11069-023-05822-4

Collections

Department of Civil Engineering, Article

Suggestions

OpenMETU
Core

Re-examination of damage distribution in Adapazari: Structural considerations Yakut, Ahmet; Bakır, Bahadır Sadık; Yilmaz, MT (2005-06-01) Adapazari was the scene of spectacular structural damage as well as widespread foundation displacement that occurred in the city during the devastating earthquake of August 17, 1999. The damage patterns observed in Adapazari were quite peculiar, so these are re-examined in an effort to answer the question of whether they are indicative of a consistent trend in terms of the building attributes and/or site conditions. For this purpose two databases comprising buildings surveyed in Adapazari after the earthqua...
Just how prescient are our building damage predictions? Gulkan, P; Bakır, Bahadır Sadık; Yakut, Ahmet; Yilmaz, MT (2003-05-13) Adapazari was the scene of spectacular structural damage as well as widespread liquefaction after the earthquake of August 17, 1999. Damage patterns observed are re-examined to investigate whether they are indicative of a consistent trend explicable in terms of the building attributes and/or site conditions. 301 buildings that had collapsed fully have been re-evaluated from their design blueprints. An examination based solely on structural attributes, including data from other sets of building assessment pr...
Fault Process and Broadband Ground-Motion Simulations of the 23 October 2011 Van (Eastern Turkey) Earthquake Gallovic, F.; Ameri, G.; Zahradnik, J.; Jansky, J.; Plicka, V.; Sokos, E.; Askan Gündoğan, Ayşegül; Pakzad, M. (2013-12-01) On 23 October 2011 an M-w 7.1 earthquake occurred in eastern Turkey, close to the towns of Van and Ercis, causing more than 600 casualties and widespread structural damage. The earthquake ruptured a 60-70 km long northeast-southwest fault with a thrust mechanism, in agreement with regional tectonic stress regime. We studied the fault process of the event and the recorded ground motions using different sets of data. Regional records (0.005-0.010 Hz) are used to constrain the centroid moment tensor solution. ...
Seismic performance of mid-rise reinforced concrete buildings in Izmir Bayrakli after the 2020 Samos earthquake Demirel, Ismail Ozan; Yakut, Ahmet; Binici, Barış (2022-07-01) © 2022 Elsevier LtdOn October 30th 2020, offshores of Samos Island, Greece was hit by a shallow earthquake of Mw = 6.9 moment magnitude. Located 76 km away from the earthquake's epicenter, the mid-rise reinforced concrete (RC) building stock of the densely populated Izmir Bayrakli district experienced the most severe structural damage. Although the horizontal peak ground acceleration recorded at the site (PGA = 0.11 g) was well below design values recommended by the past and the present Turkish seismic code...
Real-time experimental forecast of the Peruvian tsunami of August 2007 for US coastlines Wei, Yong; Bernard, Eddie N.; Tang, Liujuan; Weiss, Robert; Titov, Vasily V.; Moore, Christopher; Spillane, Michael; Hopkins, Mike; Kanoğlu, Utku (American Geophysical Union (AGU), 2008-02-27) At 23: 41 UTC on 15 August 2007, an offshore earthquake of magnitude 8.0 severely damaged central Peru and generated a tsunami. Severe shaking by the earthquake collapsed buildings throughout the region and caused 514 fatalities. The tsunami resulted in three casualties and a representative maximum runup height of similar to 7 m in the near field. The first real-time tsunami data available came from a deep-ocean tsunami detection buoy within 1 hour of tsunami generation. These tsunami data were used to prod...

Citation Formats

F. Di Michele et al., “Comparison of machine learning tools for damage classification: the case of L’Aquila 2009 earthquake,” Natural Hazards, pp. 0–0, 2023, Accessed: 00, 2023. [Online]. Available: https://hdl.handle.net/11511/102507.