Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Optimizing Parameters for Earthquake Prediction Using Bi-LSTM and Grey Wolf Optimization on Seismic Data

Rektorat UDINUS 2024 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2024
Goals
Record
905990a8-6539-49bb-b735-c5fcea308c8a
Published
2026-10-30

Universitas Dian Nuswantoro conducted an activity on optimizing parameters for earthquake prediction using a Bidirectional Long Short-Term Memory (Bi-LSTM) model and Grey Wolf Optimization (GWO) on seismic data. The activity involved Shidik, Guruh Fajar; Pramunendar, Ricardus Anggi; Purwanto; Hasibuan, Zainal Arifin; Dolphina, Erlin; Kusumawati, Yupie; and Sriwinarsih, Nurul Anisa as participants. The activity addressed the significant threat posed by earthquakes to societies worldwide and the need for advanced prediction technologies.

The activity focused on selecting suitable parameters for the Bi-LSTM model because parameter variation can affect model performance. GWO was employed to optimize the parameters, and its effectiveness was evaluated against other group optimization approaches to identify efficient parameters for earthquake prediction. A multiple input multiple output architecture was also implemented to enhance prediction accuracy. The evaluation results reported that GWO outperformed the other optimization techniques, with a reduced loss score of 0.364. The ANOVA method produced a p-value approaching 0, indicating statistical significance. In relation to the stated development goal, the activity concerns the development of early warning systems for earthquake disasters and emphasizes parameter optimization in earthquake prediction. Through this activity, UDINUS presents research focused on Bi-LSTM and GWO methodologies for earthquake prediction and their contribution to disaster early warning development.

Reference https://www.scopus.com/pages/publications/85197525425?origin=resultslist