Nonlinear Methodologies for Identifying Seismic Event and Nuclear Explosion Using Random Forest, Support Vector Machine, and Naive Bayes Classification

Abstract

The discrimination of seismic event and nuclear explosion is a complex and nonlinear system. The nonlinear methodologies including Random Forests (RF), Support Vector Machines (SVM), and Naïve Bayes Classifier (NBC) were applied to discriminant seismic events. Twenty earthquakes and twenty-seven explosions with nine ratios of the energies contained within predetermined “velocity windows” and calculated distance are used in discriminators. Based on the one out cross-validation, ROC curve, calculated accuracy of training and test samples, and discriminating performances of RF, SVM, and NBC were discussed and compared. The result of RF method clearly shows the best predictive power with a maximum area of 0.975 under the ROC among RF, SVM, and NBC. The discriminant accuracies of RF, SVM, and NBC for test samples are 92.86%, 85.71%, and 92.86%, respectively. It has been demonstrated that the presented RF model can not only identify seismic event automatically with high accuracy, but also can sort the discriminant indicators according to calculated values of weights.

Dynamic Stability Analysis of Rockmass: A Review

“…During underground activities, the accumulated elastic energy is suddenly released in the rockmasses, causing failure with a sudden. By using random forest, support vector machine, and naive Bayes classification, Dong et al obtained nonlinear methodologies for identifying the seismic event and nuclear explosion [203]. Figure 9 shows the application of ROC curves to compare the discrimination performance of different methods.…”

“…ROC of established RF, SVM (RBF), SVM (liner), and NBC models. e maximum area indicates the best predictive power[203].…”

Experimental and Numerical Evaluation of Rock Dynamic Test with Split-Hopkinson Pressure Bar

“…Since the first use of split-Hopkinson pressure bar (SHPB) system by Kolsky (1949), extensive studies have been performed to investigate dynamic mechanical properties of different materials. So far, SHPB experimental technique has been widely used in geotechnical evaluations and substantial efforts have been made to study dynamic mechanical properties of rocks [1][2][3][4][5][6][7][8][9]. The result shows that the dynamic compressive strength and dynamic tensile strength of lands measurement using SHPB are valid and reliable by Dai et al [10].…”

Theoretical and Case Studies of Interval Nonprobabilistic Reliability for Tailing Dam Stability

“…Through the introduction and intersection of some new disciplines and theories, some new methods of slope stability analysis, such as reliability analysis based on the probability theory and mathematical statistics [1,2], the comprehensive evaluation method based on the fuzzy/statistical mathematics [3][4][5][6], the gray system evaluation method based on the gray system theory [7][8][9][10][11], and the neural network evaluation method based on the neural network theory [12][13][14][15][16], are gradually formed. These evaluation methods have achieved good application in slope stability analysis and evaluation and promoted the development of slope stability research.…”