Refine
Has Fulltext
- no (79)
Year of publication
Document Type
- Article (79) (remove)
Is part of the Bibliography
- yes (79)
Keywords
- Ground-motion prediction equation (3)
- Ground-motion prediction equations (2)
- Aleatory variability (1)
- Array seismology (1)
- Attenuation (1)
- Backbone model (1)
- Bagging (1)
- Bayesian networks (1)
- Chile subduction zone (1)
- Correlation (1)
- Crustal earthquakes (1)
- Data processing (1)
- Deconvolution (1)
- Duration (1)
- Early warning (1)
- Ensembles (1)
- Epistemic uncertainty (1)
- Europe (1)
- Expert judgment (1)
- Fourier amplitude spectrum (1)
- Frequency-magnitude distribution (1)
- GPR (1)
- Ground motion prediction equation (1)
- Ground penetrating radar (1)
- Ground-motion models (1)
- Ground-motion-model (1)
- Hierarchical model (1)
- Indian Ocean (1)
- Inverse filtering (1)
- Kappa (1)
- Logic tree (1)
- Logic trees (1)
- Manifold (1)
- Middle East (1)
- Mixture model (1)
- Model selection (1)
- Multilevel model (1)
- Non-ergodic PSHA (1)
- P-waves (1)
- Probabilistic forecasting (1)
- Probabilistic seismic hazard analysis and Bayesian inference (1)
- Random vibration theory (1)
- Regional-dependence (1)
- Response Spectrum (1)
- S-waves (1)
- Seismic hazard assessment (1)
- Single-station sigma (1)
- Site amplification (1)
- Site effects (1)
- Stochastic model (1)
- Stress parameter (1)
- Strong-motion data (1)
- Style of faulting (1)
- Swarm earthquakes (1)
- Theoretical seismology (1)
- Tsunamis (1)
- Vertical resolution (1)
- Wave propagation (1)
- West Bohemia (1)
- algorithms (1)
- body waves (1)
- damage (1)
- earthquakes (1)
- elastic waves (1)
- eruptions (1)
- floods (1)
- geologic hazards (1)
- model validation (1)
- natural hazards (1)
- regression tree (1)
- seismic waves (1)
- signal-to-noise ratio (1)
- swarms (1)
- volcanic earthquakes (1)
In this article, we address the question of how observed ground-motion data can most effectively be modeled for engineering seismological purposes. Toward this goal, we use a data-driven method, based on a deep-learning autoencoder with a variable number of nodes in the bottleneck layer, to determine how many parameters are needed to reconstruct synthetic and observed ground-motion data in terms of their median values and scatter. The reconstruction error as a function of the number of nodes in the bottleneck is used as an indicator of the underlying dimensionality of ground-motion data, that is, the minimum number of predictor variables needed in a ground-motion model. Two synthetic and one observed datasets are studied to prove the performance of the proposed method. We find that mapping ground-motion data to a 2D manifold primarily captures magnitude and distance information and is suited for an approximate data reconstruction. The data reconstruction improves with an increasing number of bottleneck nodes of up to three and four, but it saturates if more nodes are added to the bottleneck.