Analysis of Seismic data using Machine Learning Algorithms. Issue 1 (February 2021)

• Record Type: Journal Article
• Authors: Sruthi, A A V L; Bhargavi, R; Gospati, Vineesha Reddy
• Is Part Of: IOP conference series. Volume 1070:Issue 1(2021)
• Publication Date: 2021-02
• Subjects: Seismic analysis -- machine learning algorithm -- Random forest algorithm -- k-NN algorithm -- earthquake magnitude -- focal depth.; Materials science -- Periodicals; 620.1105
• DOI: 10.1088/1757-899X/1070/1/012042
• Languages: English
• ISSNs: 1757-8981

Abstract

Abstract: Earthquakes result in a gigantic loss of lives and properties to people because of its powerful, devastating and deep action. Over the years, a lot of research is going on to forecast the likelihood of occurrence of an earthquake to minimize the loss. In this study, a data mining technique i.e., classification analysis has been applied to estimate the most accurate earthquake model. Previous seismic data were collected and classified by applying k-NN (k-nearest neighbors algorithm) and Random forest algorithms. k-NN is a supervised machine learning algorithm used for bigger datasets (generally for statistical estimation) to determine the accuracy of the model. Random forest algorithm is also a supervised algorithm which is used for both classification and regression. By using this algorithm, multiple decision trees can be created over the datasets as well as predicting and offering a solution. Analysis and visualization of the data has been done and subsequently a comparative analysis of these two algorithms were done and tested to obtain the efficiency in predicting the accuracy of the earthquake model in terms of earthquake magnitude and depth