Title: Associate Professor
Country/Region: Georgia
Period: 2024/9/8-2024/12/7
Theme: AI-based ground motion prediction models
Host: Hiroe MIYAKE
Introduction: Nato Jorjiashvili is an Associate Professor of geophysics at the Faculty of Natural Sciences and Medicine and simultaneously, she works as a Consultant at the Institute of Earth Sciences, and National Seismic Monitoring Centre, Ilia State University, Georgia. Her work focusses on using modern technologies such as Machine Learning and Artificial Intelligence in seismology, namely for ground motion prediction models and seismic hazard assessment. Her work is also devoted to assessing model reliability and validity. In addition, one of her research topics is the assessment of uncertainty related to models, especially the influence of uncertainty and error of each parameter involved in the process of seismic hazard assessment on the result. She also examines how the study results are improved by the use of modern technologies such as machine learning and artificial intelligence.
Research Report:
With the support of ERI and associate professor Hiroe Miyake I had a great chance to visit ERI and conduct interesting research. I have worked on ground motion prediction models (GMPM) using artificial neural networks (ANN). To find out differences in regions and benefit of the method I have used two datasets: 1) One for Japan (data from NIED - National Research Institute for Earth Science and Disaster Resilience) (Fig.1); 2) Data from Georgia (Institute of Earth Sciences and National Seismic Monitoring Centre, Ilia State University) (Fig.2). In case of Georgia due to the lack of enough strong motion records I have used hybrid dataset strong motion type together with broadband type data.
View All
Fiscal Year: 2024
Fiscal Year: 2024