Using Machine Learning Models for Earthquake Magnitude Prediction in California, Japan, and Israel

Deborah Novick; Mark Last

doi:10.1007/978-3-031-34671-2_11

Using Machine Learning Models for Earthquake Magnitude Prediction in California, Japan, and Israel

Deborah Novick
, Mark Last

The Goldman Sonnenfeldt School of Sustainability and Climate Change

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

5 Scopus citations

Abstract

This study aims at predicting whether an earthquake of magnitude greater than the regional median of maximum yearly magnitudes will occur during the next year. Prediction is performed by training various machine learning algorithms, such as AdaBoost, XGBoost, Random Forest, Logistic Regression, and Info-Fuzzy Network. The models are induced using a combination of seismic indicators used in the earthquake literature as well as various time-series features, such as features based on the moving averages of the number of earthquakes in each area, features that record the number of events above and below the mean in a time period, and features based on lagged values of the mean and median magnitude. Feature selection is performed using a forward search algorithm that chooses the most effective features for prediction. The models are trained and evaluated using earthquake catalog data obtained for California, Japan, and Israel. In addition, models trained on either California or Japan datasets are evaluated using the remaining data. Models trained on Japan data achieve AUC scores up to 0.825; models trained on California data achieve AUC scores up to 0.738; and models trained on Israel data achieve AUC scores up to 0.710.

Original language	English
Title of host publication	Cyber Security, Cryptology, and Machine Learning - 7th International Symposium, CSCML 2023, Proceedings
Editors	Shlomi Dolev, Ehud Gudes, Pascal Paillier
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	151-169
Number of pages	19
ISBN (Print)	9783031346705
DOIs	https://doi.org/10.1007/978-3-031-34671-2_11
State	Published - 1 Jan 2023
Event	7th International Symposium on Cyber Security, Cryptology, and Machine Learning, CSCML 2023 - Be'er Sheva, Israel Duration: 29 Jun 2023 → 30 Jun 2023

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13914 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	7th International Symposium on Cyber Security, Cryptology, and Machine Learning, CSCML 2023
Country/Territory	Israel
City	Be'er Sheva
Period	29/06/23 → 30/06/23

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 11 Sustainable Cities and Communities

Keywords

Classification models
Clustering analysis
Earthquake prediction
Seismicity indicators

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-031-34671-2_11

Cite this

Novick, D., & Last, M. (2023). Using Machine Learning Models for Earthquake Magnitude Prediction in California, Japan, and Israel. In S. Dolev, E. Gudes, & P. Paillier (Eds.), Cyber Security, Cryptology, and Machine Learning - 7th International Symposium, CSCML 2023, Proceedings (pp. 151-169). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13914 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-34671-2_11

Novick, Deborah ; Last, Mark. / Using Machine Learning Models for Earthquake Magnitude Prediction in California, Japan, and Israel. Cyber Security, Cryptology, and Machine Learning - 7th International Symposium, CSCML 2023, Proceedings. editor / Shlomi Dolev ; Ehud Gudes ; Pascal Paillier. Springer Science and Business Media Deutschland GmbH, 2023. pp. 151-169 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{17141a2f6b8346b1b3146c73bfdd579d,

title = "Using Machine Learning Models for Earthquake Magnitude Prediction in California, Japan, and Israel",

abstract = "This study aims at predicting whether an earthquake of magnitude greater than the regional median of maximum yearly magnitudes will occur during the next year. Prediction is performed by training various machine learning algorithms, such as AdaBoost, XGBoost, Random Forest, Logistic Regression, and Info-Fuzzy Network. The models are induced using a combination of seismic indicators used in the earthquake literature as well as various time-series features, such as features based on the moving averages of the number of earthquakes in each area, features that record the number of events above and below the mean in a time period, and features based on lagged values of the mean and median magnitude. Feature selection is performed using a forward search algorithm that chooses the most effective features for prediction. The models are trained and evaluated using earthquake catalog data obtained for California, Japan, and Israel. In addition, models trained on either California or Japan datasets are evaluated using the remaining data. Models trained on Japan data achieve AUC scores up to 0.825; models trained on California data achieve AUC scores up to 0.738; and models trained on Israel data achieve AUC scores up to 0.710.",

keywords = "Classification models, Clustering analysis, Earthquake prediction, Seismicity indicators",

author = "Deborah Novick and Mark Last",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 7th International Symposium on Cyber Security, Cryptology, and Machine Learning, CSCML 2023 ; Conference date: 29-06-2023 Through 30-06-2023",

year = "2023",

month = jan,

day = "1",

doi = "10.1007/978-3-031-34671-2\_11",

language = "English",

isbn = "9783031346705",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "151--169",

editor = "Shlomi Dolev and Ehud Gudes and Pascal Paillier",

booktitle = "Cyber Security, Cryptology, and Machine Learning - 7th International Symposium, CSCML 2023, Proceedings",

address = "Germany",

}

Novick, D & Last, M 2023, Using Machine Learning Models for Earthquake Magnitude Prediction in California, Japan, and Israel. in S Dolev, E Gudes & P Paillier (eds), Cyber Security, Cryptology, and Machine Learning - 7th International Symposium, CSCML 2023, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13914 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 151-169, 7th International Symposium on Cyber Security, Cryptology, and Machine Learning, CSCML 2023, Be'er Sheva, Israel, 29/06/23. https://doi.org/10.1007/978-3-031-34671-2_11

Using Machine Learning Models for Earthquake Magnitude Prediction in California, Japan, and Israel. / Novick, Deborah; Last, Mark.
Cyber Security, Cryptology, and Machine Learning - 7th International Symposium, CSCML 2023, Proceedings. ed. / Shlomi Dolev; Ehud Gudes; Pascal Paillier. Springer Science and Business Media Deutschland GmbH, 2023. p. 151-169 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13914 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Using Machine Learning Models for Earthquake Magnitude Prediction in California, Japan, and Israel

AU - Novick, Deborah

AU - Last, Mark

PY - 2023/1/1

Y1 - 2023/1/1

N2 - This study aims at predicting whether an earthquake of magnitude greater than the regional median of maximum yearly magnitudes will occur during the next year. Prediction is performed by training various machine learning algorithms, such as AdaBoost, XGBoost, Random Forest, Logistic Regression, and Info-Fuzzy Network. The models are induced using a combination of seismic indicators used in the earthquake literature as well as various time-series features, such as features based on the moving averages of the number of earthquakes in each area, features that record the number of events above and below the mean in a time period, and features based on lagged values of the mean and median magnitude. Feature selection is performed using a forward search algorithm that chooses the most effective features for prediction. The models are trained and evaluated using earthquake catalog data obtained for California, Japan, and Israel. In addition, models trained on either California or Japan datasets are evaluated using the remaining data. Models trained on Japan data achieve AUC scores up to 0.825; models trained on California data achieve AUC scores up to 0.738; and models trained on Israel data achieve AUC scores up to 0.710.

AB - This study aims at predicting whether an earthquake of magnitude greater than the regional median of maximum yearly magnitudes will occur during the next year. Prediction is performed by training various machine learning algorithms, such as AdaBoost, XGBoost, Random Forest, Logistic Regression, and Info-Fuzzy Network. The models are induced using a combination of seismic indicators used in the earthquake literature as well as various time-series features, such as features based on the moving averages of the number of earthquakes in each area, features that record the number of events above and below the mean in a time period, and features based on lagged values of the mean and median magnitude. Feature selection is performed using a forward search algorithm that chooses the most effective features for prediction. The models are trained and evaluated using earthquake catalog data obtained for California, Japan, and Israel. In addition, models trained on either California or Japan datasets are evaluated using the remaining data. Models trained on Japan data achieve AUC scores up to 0.825; models trained on California data achieve AUC scores up to 0.738; and models trained on Israel data achieve AUC scores up to 0.710.

KW - Classification models

KW - Clustering analysis

KW - Earthquake prediction

KW - Seismicity indicators

UR - https://www.scopus.com/pages/publications/85164965545

U2 - 10.1007/978-3-031-34671-2_11

DO - 10.1007/978-3-031-34671-2_11

M3 - Conference contribution

AN - SCOPUS:85164965545

SN - 9783031346705

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 151

EP - 169

BT - Cyber Security, Cryptology, and Machine Learning - 7th International Symposium, CSCML 2023, Proceedings

A2 - Dolev, Shlomi

A2 - Gudes, Ehud

A2 - Paillier, Pascal

PB - Springer Science and Business Media Deutschland GmbH

T2 - 7th International Symposium on Cyber Security, Cryptology, and Machine Learning, CSCML 2023

Y2 - 29 June 2023 through 30 June 2023

ER -

Novick D, Last M. Using Machine Learning Models for Earthquake Magnitude Prediction in California, Japan, and Israel. In Dolev S, Gudes E, Paillier P, editors, Cyber Security, Cryptology, and Machine Learning - 7th International Symposium, CSCML 2023, Proceedings. Springer Science and Business Media Deutschland GmbH. 2023. p. 151-169. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-34671-2_11