Treffer: A Systematic Review About the Use of Machine Learning Related to Earthquake Studies.

This systematic review explores the application of machine learning (ML) techniques in earthquake prediction, analyzing studies published between 2018 and 2022. The research focuses on identifying models, methods, and tools used in this field, as well as evaluating their effectiveness. A systematic methodology based on Kitchenham's framework was employed, including three main phases: planning, conducting, and reporting the review. The process involved formulating research questions (RQs), rigorously searching 11 academic databases, and applying inclusion and exclusion criteria to refine 56,240 initial records into 126 relevant studies. Key methods identified include supervised, unsupervised, and reinforcement learning models, with supervised learning being the most utilized approach. Prominent techniques include Naive Bayes (NB), K‐means, lasso regression, ridge regression, and random forest (RF). Variables frequently associated with earthquake prediction include seismic precursors, neural networks, and prediction accuracy metrics. Python and TensorFlow were the most commonly used tools for implementing these methods. The findings reveal that while ML holds significant potential for improving earthquake prediction, current research is predominantly focused on supervised learning, with limited exploration of other methodologies. The review highlights the need for diverse approaches and further evaluation of underutilized techniques, emphasizing their importance for advancing predictive models. This work contributes a comprehensive analysis of the current state of ML in seismic studies, identifying gaps and opportunities for future research. [ABSTRACT FROM AUTHOR]

Copyright of Advances in Civil Engineering is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)