Application and visualization of typical clustering algorithms in seismic data analysis
暂无分享,去创建一个
Abstract Earthquake brings enormous loss of lives and properties to human beings due to its suddenness, destructiveness and inscrutability. The new techniques for analyzing seismic data can reveal the distribution of earthquakes, which helps us master the laws of earthquake disasters and reduce the risks brought by them. In this paper, we applied K-means and DBSCAN clustering algorithms to the analysis of seismic data. Their performances in fitting seismic belts with seismic datasets are compared. First, we map the positional parameters in the seismic data to coordinate points on a two-dimensional plane and then cluster them with the DBSCAN algorithm. In addition, we combine the magnitude and depth properties of seismic data, use the Elbow method to find the best K value, and then classifies the dataset by K-means algorithm. We visualize the results, and the distinction of each classification is clear. The experimental results show that the DBSCAN algorithm has a better effect on fitting the seismic belt, and the classification results of K-means algorithm for earthquakes are also in line with expectations.
[1] Pan Ruo-yu,et al. Optimization Study on k Value of K-means Algorithm , 2006 .
[2] Han Peng,et al. Principal component analysis of geomagnetic diurnal variation associated with earthquakes:case study of the M6.1 Iwate-ken Nairiku Hokubu earthquake , 2009 .
[3] Xiaolu Li,et al. Fractal dimension analysis for seismicity spatial and temporal distribution in the circum-Pacific seismic belt , 2019, Journal of Earth System Science.