Intelligent Methods with Applications in Volcanology and Seismology

Preface
Reference
Contents
1 Intelligent Methods and Motivations to Use in Volcanology and Seismology
Abstract
1 Introduction
1.1 Brief List of Intelligent Methods Applications in Volcanology and Seismology
1.1.1 In Volcanology
1.1.2 In Seismology
1.2 Metaheuristic Algorithms
1.3 Intelligent Methods
1.3.1 Motivations
1.3.2 Machine Learning; an Overview
1.4 The Role of Intelligent Methods Toward Big Volcano Science
References
2 Machine Learning: The Concepts
Abstract
1 Introduction
2 An Overview of the State Estimation Problem
3 Parametric and Non-parametric Estimation of Densities
3.1 A Parametric Approach
3.2 Parametric Density Estimation: A Numerical Example
3.3 A Non-parametric Approach
3.4 Estimation of the Prior Probabilities
4 Supervised Classification
4.1 The Bayes Minimum Risk Classification
4.2 An Example of Bayesian Minimum Risk Classifier
4.3 Naive Bayes Classifiers
4.4 Parzen Classifiers
4.5 K-Nearest Neighbor (KNN) Classification
4.6 Classification Based on Discriminant Functions
4.7 The Support Vector Classifier
4.8 Decision Trees
4.9 Combining Models: Boosting and Bagging
4.9.1 Boosting
4.9.2 Bagging
4.10 Error-Correcting Output Codes (ECOC)
4.11 Hidden Markov Models
5 Classification Metrics for Model Validation
6 Unsupervised Classification
6.1 Hierarchical Clustering
6.2 K-Means Clustering
6.3 Fuzzy c-means
6.4 Mixture of Gaussians
7 Methods to Reduce the Dimensionality of a Dataset
7.1 The Principal Component Analysis (PCA)
7.2 Self-organizing Maps
8 Software Tools for Machine Learning
8.1 The MATLAB™ Statistical and Machine Learning Toolbox
8.2 The Python Scikit-Learn Package
8.3 The R Language
8.4 The PRTools Library
References
3 Machine Learning Applications in Volcanology and Seismology
Abstract
1 Introduction
2 ML to Classify Seismic Data
3 Hidden Markov Model to Classify Volcanic Activity
4 Earthquake Detection and Phase Picking
5 Earthquake and Early Warning
6 Ground Motion Prediction
7 ML for Volcanic Activity Monitoring Based on Images
8 Multi-parametric Approaches to Classify the Volcanic Activity
9 Unsupervised Classification of Volcanic Activity
10 Clustering Multivariate Geophysical Data by Using SOM
References
4 Deep Learning: The Concepts
Abstract
1 Introduction
2 Deep Learning, an Overview
3 Deep Learning, Pros and Cons
4 Layers in a Deep Learning Models
5 Deep Learning Models
5.1 Supervised Deep Learning Methods
5.2 Deep Convolutional Neural Network
5.3 Image Classification by CNN: Fault Detection in Synthetic Seismic Data
5.4 Recurrent Neural Networks
5.5 Long Short Term Memory Network
5.6 Gated Recurrent Unit Network
5.7 Application of Long Short Term Memory Network for Extrapolating 2D Sequential Data
5.8 Unsupervised Deep Learning Methods
5.9 Unsupervised Auto Encoder Network
5.10 Attenuating Random Noise in Gravity Data Using Auto Encoder Network
5.10.1 Generative Adversarial Network
References
5 Deep Learning: Applications in Seismology and Volcanology
Abstract
1 Introduction
2 Applications of Deep Learning in Seismology
2.1 Long-Range-Short-Term Earthquake Prediction Using CNN-BiLSTM-AM Model
2.2 Real-Time Focal Mechanism Determination by Fully Convolutional Network
2.3 A Functional Very Deep Convolutional Neural Network Model for Fast and Precise Earthquake Phase Picking
2.4 Magnitude Calculation Directly from Raw Waveforms Using MagNet
3 Detecting and Locating Induced Seismicity Using ConvNetQuake Model
3.1 Classification Based on Limited Training Samples Using CapsNet: Aplication to Microseismic Record Classification
3.2 Deep Convolutional Neural Network for Fast Prediction of Earthquake Intensity from Raw Accelerograms
4 Applications of Deep Learning in Volcanology
4.1 Volcano Deformation Identification Using Convolutional Neural Network Trained by InSAR Synthetic Database
4.2 Automatic Classification of Volcano-Seismic Signals Using Active Deep Learning to Overcome the Case of Small Training Datasets
4.3 Probabilistic Shape Classification of the Volcanic Ash Using Convolutional Neural Networks
References
6 Evolutionary Algorithms with Focus on Genetic Algorithm
Abstract
1 Evolutionary Computation
1.1 Introduction
1.2 Annals of Evolutionary Computing: A Brief Literature Review
1.3 Biological and Artificial Evolutionary
2 Genetic Algorithm
2.1 Introduction to Natural Genetics
2.2 Crossover
2.3 Mutation
2.4 Fitness
3 Fundamentals of Genetic Algorithms
3.1 Encoding
3.2 Fitness
3.3 Cross Over
3.4 Mutation
4 How to Run Each Parts of the Genetic Algorithm?
4.1 Population Representation and Initializing the Algorithm
4.2 Objective Function and Fitting Function
4.3 Selection and Various Techniques
4.4 Different Techniques of Cross Over
4.5 Different Techniques of Mutation
5 Important Definitions in Running Genetic Algorithm
6 The General Process of Optimization and Problem Solving in Genetic Algorithms
7 More Examples on Operators in Genetic Algorithms
7.1 Encoding a Chromosome
7.2 Cross Over
7.3 Mutation
8 Investigation of Important Factors in Genetic Algorithm
8.1 Cross Over Rate
8.2 Mutation Rate
8.3 Population Size
8.4 Selection Methods
8.5 Different Types of Encoding
8.5.1 Permutation Encoding
8.5.2 Value Encoding
8.5.3 Tree Encoding and Addressing
9 Genetic Algorithm in MATLAB; A Brief View
10 Random Numbers Generation in Matlab
10.1 Random Permutation
10.2 Pseudorandom Integers from a Uniform Discrete Distribution
10.3 Normally Distributed Pseudorandom Numbers
References
7 Application of Genetic Algorithm in Volcanology and Seismology
Abstract
1 Introduction
2 Inverse Modelling of Volcanomagnetic Fields Using Genetic Algorithm
2.1 Mechanisms that Cause Volcanomagnetic Anomalies
2.2 Motivations to Use Genetic Algorithm for Inversion of Volcanomagnetic Anomalies
2.3 The GA Procedure to Invert Volcanomagnetic Anomalies
2.4 Forward Models
2.4.1 Piezomagnetic Field Forward Model
2.4.2 Electrokinetic Effects Forward Model
2.4.3 Thermomagnetic Phenomena Forward Model
2.5 Testing and Evaluating the GA Performance for Synthetic Data
2.6 Evaluation of GA Results for Real Cases
3 Inversion of SAR Data in Active Volcanic Areas by Genetic Algorithm
3.1 Abstract
3.2 SAR Data Forward Modelling
3.3 Test of GA Method for Synthetic Data
3.4 Inversion of Real SAR Data
4 Automatic Monitoring System of Infrasonic Events at Mt. Etna Using Genetic Approach
4.1 The Infrasonic Events Automatic Monitoring
4.2 Genetic Algorithm Method for Infrasonic Source Parameters Estimation
4.3 Evaluation of the Proposed Method
5 Rapid Estimation of Earthquake Magnitude and Source Parameters Using Genetic Algorithms
5.1 Displacement Detection and Estimation
5.2 Moment Magnitude (Mw) Estimation
5.2.1 GA-Okada Procedure
5.2.2 Evaluation for Real Data
6 Generator of Genetic Seismic Signals
6.1 Introduction
6.2 The Underlying Idea
6.3 Evaluation of GA Seismic Generator
7 Focal-Mechanism Determination in Taiwan by Genetic Algorithm
7.1 The Study Region
7.2 GA Procedure for Focal Mechanism Determination
7.3 Test of GA for Synthetic Data
7.4 Test for Real Data
References