Feasibility of principal component analysis for multi-class earthquake prediction machine learning model utilizing geomagnetic field data
Geomagnetic field data have been found to contain earthquake (EQ) precursory signals; however, analyzing this high-resolution, imbalanced data presents challenges when implementing machine learning (ML). This study explored feasibility of principal component analyses (PCA) for reducing the dimension...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Multidisciplinary Digital Publishing Institute
2024
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/113525/ http://psasir.upm.edu.my/id/eprint/113525/1/113525.pdf |
| _version_ | 1848866250756718592 |
|---|---|
| author | Qaedi, Kasyful Abdullah, Mardina Yusof, Khairul Adib Hayakawa, Masashi |
| author_facet | Qaedi, Kasyful Abdullah, Mardina Yusof, Khairul Adib Hayakawa, Masashi |
| author_sort | Qaedi, Kasyful |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | Geomagnetic field data have been found to contain earthquake (EQ) precursory signals; however, analyzing this high-resolution, imbalanced data presents challenges when implementing machine learning (ML). This study explored feasibility of principal component analyses (PCA) for reducing the dimensionality of global geomagnetic field data to improve the accuracy of EQ predictive models. Multi-class ML models capable of predicting EQ intensity in terms of the Mercalli Intensity Scale were developed. Ensemble and Support Vector Machine (SVM) models, known for their robustness and capabilities in handling complex relationships, were trained, while a Synthetic Minority Oversampling Technique (SMOTE) was employed to address the imbalanced EQ data. Both models were trained on PCA-extracted features from the balanced dataset, resulting in reasonable model performance. The ensemble model outperformed the SVM model in various aspects, including accuracy (77.50% vs. 75.88%), specificity (96.79% vs. 96.55%), F1-score (77.05% vs. 76.16%), and Matthew Correlation Coefficient (73.88% vs. 73.11%). These findings suggest the potential of a PCA-based ML model for more reliable EQ prediction. |
| first_indexed | 2025-11-15T14:17:37Z |
| format | Article |
| id | upm-113525 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T14:17:37Z |
| publishDate | 2024 |
| publisher | Multidisciplinary Digital Publishing Institute |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-1135252024-11-26T03:20:46Z http://psasir.upm.edu.my/id/eprint/113525/ Feasibility of principal component analysis for multi-class earthquake prediction machine learning model utilizing geomagnetic field data Qaedi, Kasyful Abdullah, Mardina Yusof, Khairul Adib Hayakawa, Masashi Geomagnetic field data have been found to contain earthquake (EQ) precursory signals; however, analyzing this high-resolution, imbalanced data presents challenges when implementing machine learning (ML). This study explored feasibility of principal component analyses (PCA) for reducing the dimensionality of global geomagnetic field data to improve the accuracy of EQ predictive models. Multi-class ML models capable of predicting EQ intensity in terms of the Mercalli Intensity Scale were developed. Ensemble and Support Vector Machine (SVM) models, known for their robustness and capabilities in handling complex relationships, were trained, while a Synthetic Minority Oversampling Technique (SMOTE) was employed to address the imbalanced EQ data. Both models were trained on PCA-extracted features from the balanced dataset, resulting in reasonable model performance. The ensemble model outperformed the SVM model in various aspects, including accuracy (77.50% vs. 75.88%), specificity (96.79% vs. 96.55%), F1-score (77.05% vs. 76.16%), and Matthew Correlation Coefficient (73.88% vs. 73.11%). These findings suggest the potential of a PCA-based ML model for more reliable EQ prediction. Multidisciplinary Digital Publishing Institute 2024 Article PeerReviewed text en cc_by_4 http://psasir.upm.edu.my/id/eprint/113525/1/113525.pdf Qaedi, Kasyful and Abdullah, Mardina and Yusof, Khairul Adib and Hayakawa, Masashi (2024) Feasibility of principal component analysis for multi-class earthquake prediction machine learning model utilizing geomagnetic field data. Geosciences, 14 (5). art. no. 121. pp. 1-11. ISSN 2076-3263; eISSN: 2076-3263 https://www.mdpi.com/2076-3263/14/5/121 10.3390/geosciences14050121 |
| spellingShingle | Qaedi, Kasyful Abdullah, Mardina Yusof, Khairul Adib Hayakawa, Masashi Feasibility of principal component analysis for multi-class earthquake prediction machine learning model utilizing geomagnetic field data |
| title | Feasibility of principal component analysis for multi-class earthquake prediction machine learning model utilizing geomagnetic field data |
| title_full | Feasibility of principal component analysis for multi-class earthquake prediction machine learning model utilizing geomagnetic field data |
| title_fullStr | Feasibility of principal component analysis for multi-class earthquake prediction machine learning model utilizing geomagnetic field data |
| title_full_unstemmed | Feasibility of principal component analysis for multi-class earthquake prediction machine learning model utilizing geomagnetic field data |
| title_short | Feasibility of principal component analysis for multi-class earthquake prediction machine learning model utilizing geomagnetic field data |
| title_sort | feasibility of principal component analysis for multi-class earthquake prediction machine learning model utilizing geomagnetic field data |
| url | http://psasir.upm.edu.my/id/eprint/113525/ http://psasir.upm.edu.my/id/eprint/113525/ http://psasir.upm.edu.my/id/eprint/113525/ http://psasir.upm.edu.my/id/eprint/113525/1/113525.pdf |