Affiliation:
1. York St. John University, UK
Abstract
An earthquake is one of the deadliest natural disasters. Forecasting an earthquake is a challenging task since natural causes such as rainfall or volcanic eruptions disrupt data. Earthquakes can also be caused by human beings, such as mining or dams. Solar activity has also been suggested as a possible cause of earthquakes. Solar activity and earthquakes occur in different parts of the solar system, separated by a huge distance. However, scientists have been trying to figure out if there are any links between these two seemingly unrelated occurrences since the 19th century. In this chapter, the authors explored the methods of how machine learning algorithms including k-nearest neighbour, support vector regression, random forest regression, and long short-term memory neural networks can be applied to predict earthquakes and to understand if there is a relationship between solar activity and earthquakes. The authors investigated three types of solar activity: sunspots number, solar wind, and solar flares, as well as worldwide earthquake frequencies that ranged in magnitude and depth.
Reference91 articles.
1. AlpaydinE. (2014). Introduction to machine learning (3rd ed.). The MIT Press.
2. Using support vector machine (SVM) and Ionospheric Total Electron Content (TEC) data for solar flare predictions.;S.Asaly;IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2021
3. Earthquake magnitude prediction in Hindukush region using machine learning techniques.;K. M.Asim;Natural Hazards,2017
4. Health assessment and life prediction of cutting tools based on support vector regression.;T.Benkedjouh;Journal of Intelligent Manufacturing,2015
5. The effect of solar cycle’s activities on earthquake: A conceptual idea for forecasting.;N.Bijan;Disaster Advances,2013