We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
A systematic review of Earthquake Early Warning (EEW) systems based on Artificial Intelligence.
- Authors
Kolivand, Pirhossein; Saberian, Peyman; Tanhapour, Mozhgan; Karimi, Fereshteh; Kalhori, Sharareh Rostam Niakan; Javanmard, Zohreh; Heydari, Soroush; Talari, Seyed Saeid Hoseini; Mousavi, Seyed Mohsen Laal; Alidadi, Maryam; Ahmadi, Mahnaz; Ayyoubzadeh, Seyed Mohammad
- Abstract
Early Earthquake Warning (EEW) systems alarm about ongoing earthquakes to reduce their devastating human and financial damages. In complicated tasks like earthquake forecasting, Artificial Intelligence (AI) solutions show promising results. The goal of this review is to investigate the AI-based EEW systems. Web of Science, Scopus, Embase, and PubMed databases were systematically searched from its beginning until April 18, 2023. Studies that used AI algorithms to develop EEWs and forecast earthquake magnitude were qualified. The quality assessment was conducted using the Mixed Methods Assessment Tool version 2018. Detailed analysis was performed on 26 of 2604 retrieved articles. Researchers predict earthquakes most often using neural network family models (21 studies). Among eight categorized groups of parameters for earthquake forecasting, it was often predicted utilizing seismic wave characteristics (65.38%) and seismic activity data (61.54%). AI models most often predicted earthquake magnitude (32.69%) and depth (15.38%). Logistic Model Tree and Bayesian Network had the highest sensitivity, accuracy, and F-measure efficiency (99.9%). Findings showed that AI algorithms can forecast earthquakes. However, additional study is needed to determine the efficacy of more data-driven AI algorithms in mining seismic data using more input variables. This review is helpful for seismologists and researchers developing EEW systems using AI.
- Subjects
ARTIFICIAL intelligence; EARTHQUAKES; EARTHQUAKE prediction; EARTHQUAKE magnitude; SEISMIC waves
- Publication
Earth Science Informatics, 2024, Vol 17, Issue 2, p957
- ISSN
1865-0473
- Publication type
Article
- DOI
10.1007/s12145-024-01253-2