Search

The presence of "ill-posed samples" specifically in low-volume datasets leads to accuracy decrement in the learning procedure and the generalization of neural networks. Such samples can be caused by various reasons such as noise contamination, corrupted sensors, or even, the complex distribution of physical properties governing the problem. The peak ground acceleration (PGA) datasets are definitely among the last mentioned. Focusing on speed and accuracy, a method for calculating earthquake magnitude based on the PGA data recorded at a single station along with hypocentral information has been presented in this research. Here, after training a deep neural network, the regression errors of the training data samples are clustered into two groups, namely well and ill posed using the grey wolf optimization algorithm. Instead of being removed, the data samples with low learning rates are then modified using samples selected from the other cluster in a fusional form. Then, two separate models are used and trained independently for the clusters. Next, in addition to the routine procedure of network generalization, every new sample is first checked whether is more likely to belong to which group of the clustered data, and after processing, the corresponding trained model is used. The results of the experiments show that using the proposed method results in magnitude calculation with an error order of less than 0.212 units of moment magnitude with a probability of more than 99.7%, which is superior to the conventional methods some of which were reviewed in this research. [ABSTRACT FROM AUTHOR]

In this paper, a method is presented to classify volcano activity into three classes, namely quiet, strombolian, and paroxysm. The method is based on training a six-layered deep neural network (DNN) model using these signals as inputs (features): time series of the number of distances of infrasonic events, radar backscatter power, RMS of tremor in five stations close to craters of the volcano, tilt derivative, and seismic tremor source depth. The method was tested on the data related to a period of five years, and the results were concluded using indexes of precision, recall, F1 score, and Cohen's Kappa coefficient were calculated to evaluate the qualification of the classification. Also, the results were compared to Bayesian network (BN), K-nearest neighbors (KNN), and decision tree (DT) methods. Decision learning trees and KNN are popular machine learning algorithms belonging to the class of supervised learning algorithms. They mimic the human level thinking and, differing from neural networks, are not black box models. The comparisons reveal the proposed method, especially in classifying both strombolian and paroxysm classes. This advantage makes the presented method a more reliable tool for practical use in the volcano monitoring control rooms. [ABSTRACT FROM AUTHOR]