Scalable earthquake magnitude prediction using spatio-temporal data and model versioning

The objective of this paper is to develop a predictive model for earthquake magnitude using machine learning techniques by integrating seismic and environmental data. This approach aims to address the challenges of earthquake prediction-such as data scarcity, regional variability, and model generalizability-by leveraging large-scale datasets and MLOps practices to enhance model robustness, accuracy, and adaptability.

The results indicate that Gradient Boosting has a strong performance on smaller datasets, XGBoost achieves the best scores with medium-sized datasets, while LightGBM provides optimal performances using the whole dataset. These changing characteristics emphasize that optimal predictions can only be obtained using dynamic algorithm selection, given the nature of the data at hand.