Open Access

ARTICLE

Error Analysis of Geomagnetic Field Reconstruction Model Using Negative Learning for Seismic Anomaly Detection

Nur Syaiful Afrizal¹, Khairul Adib Yusof^1,2,*, Lokman Hakim Muhamad¹, Nurul Shazana Abdul Hamid^2,3, Mardina Abdullah^2,4, Mohd Amiruddin Abd Rahman¹, Syamsiah Mashohor⁵, Masashi Hayakawa^6,7

1 Department of Physics, Faculty of Science, Universiti Putra Malaysia, Seri Kembangan, 43400, Malaysia
2 Space Science Center, Institute of Climate Change, Universiti Kebangsaan Malaysia, Bangi, 43600, Malaysia
3 Department of Applied Physics, Faculty of Science and Technology, Universiti Kebangsaan Malaysia, Bangi, 43600, Malaysia
4 Department of Electrical, Electronic and Systems Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, Bangi, 43600, Malaysia
5 Department of Computer and Communication System Engineering, Faculty of Engineering, Universiti Putra Malaysia, Seri Kembangan, 43400, Malaysia
6 UEC Alliance Center, Hayakawa Institute of Seismo Electromagnetics Co., Ltd. (Hi-SEM), Chofu, Tokyo, 1820026, Japan
7 Advanced Wireless & Communications Research Center (AWCC), University of Electro-Communications, Chofu, Tokyo, 182-8585, Japan

* Corresponding Author: Khairul Adib Yusof. Email:

(This article belongs to the Special Issue: Advances in Pattern Recognition Applications)

Computers, Materials & Continua 2026, 86(2), 1-16. https://doi.org/10.32604/cmc.2025.066421

Received 08 April 2025; Accepted 21 August 2025; Issue published 09 December 2025

Abstract

Detecting geomagnetic anomalies preceding earthquakes is a challenging yet promising area of research that has gained increasing attention in recent years. This study introduces a novel reconstruction-based modeling approach enhanced by negative learning, employing a Bidirectional Long Short-Term Memory (BiLSTM) network explicitly trained to accurately reconstruct non-seismic geomagnetic signals while intentionally amplifying reconstruction errors for seismic signals. By penalizing the model for accurately reconstructing seismic anomalies, the negative learning approach effectively magnifies the differences between normal and anomalous data. This strategic differentiation enhances the sensitivity of the BiLSTM network, enabling improved detection of subtle geomagnetic anomalies that may serve as earthquake precursors. Experimental validation clearly demonstrated statistically significant higher reconstruction errors for seismic signals compared to non-seismic signals, confirmed through the Mann-Whitney U test with a p-value of 0.0035 for Root Mean Square Error (RMSE). These results provide compelling evidence of the enhanced anomaly detection capability achieved through negative learning. Unlike traditional classification-based methods, negative learning explicitly encourages sensitivity to subtle precursor signals embedded within complex geomagnetic data, establishing a robust basis for further development of reliable earthquake prediction methods.

---

Keywords

---