Deep Multi-scale Dilated Convolution Neural Network with Attention Mechanism: A Novel Method for Earthquake Magnitude Classification

Kavianpour, Parisa; Kavianpour, Mohammadreza; Ramezani, Amin

doi:10.1109/icspis56952.2022.10043978

Cited by 4 publications

(2 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The optimal values of µ = 1 and γ = 0.05 are determined through experimental investigations. In the training phase, the network's parameters are initialized using the Xavier initializer [52], and the Adam optimizer [53] is employed for the training process. The datasets are partitioned into training, validation, and test sets, with proportions of 80%, 10%, and 10%, respectively.…”

Section: Implementation Detailsmentioning

confidence: 99%

A class alignment method based on graph convolution neural network for bearing fault diagnosis in presence of missing data and changing working conditions

Kavianpour

Ramezani²,

Beheshti³

2022

Measurement

View full text Add to dashboard Cite

Section: Implementation Detailsmentioning

confidence: 99%

A class alignment method based on graph convolution neural network for bearing fault diagnosis in presence of missing data and changing working conditions

Kavianpour

Ramezani²,

Beheshti³

2022

Measurement

View full text Add to dashboard Cite

“…DL algorithm has complex structure consists of input layer with their own feature extraction layer, multiple hidden layers constructed from dense layer, and connected neurons which has high generalization power [12]. This structure has increased the learning capability of the model compared to superficial neural network models [13]. In term of earthquakes prediction, Berhich et al [14] proposed the implementation of improved RNN (recurrent neural network) approaches which are long short-term memory (LSTM), gate recurrent network (GRU) and the fusion LSTM-GRU.…”

Section: Introductionmentioning

confidence: 99%

LUTanh Activation Function to Optimize BI-LSTM in Earthquake Forecasting

2024

IJIES

View full text Add to dashboard Cite

Earthquakes, as unpredictable and potentially catastrophic events, have long captured the attention of geophysicists due to their profound impact on communities. The devastating consequences of these events underscore the critical need for early earthquake prediction systems capable of forecasting location, magnitude, and depth. With the rapid advancements in technology, particularly in the fields of data science, earthquake prediction methods have undergone significant evolution. This study introduces a proposed method that modified ReLU (Rectified Linear Unit) called LUTanh (Linear Unit Hyperbolic Tangent) which combine both benefits of ReLU and Tanh activation functions. This study applied and compared the proposed method performance in long short-term memory (LSTM) and Bidirectional LSTM (Bi-LSTM) algorithms for predicting earthquake disaster. Furthermore, this proposed method was tested in architecture of multiple input multiple output (MIMO) principle in predicting the occurrence of earthquake disaster. The evaluation results decisively reveal that the LUTanh applied in Bi-LSTM model, particularly when optimized with the Adam optimizer, consistently outperforms the LSTM counterpart. Error assessments of LUTanh in Bi-LSTM consistently demonstrate lower average error scores compared to the origin ReLU activation function up to 4% of mean absolute error (MAE) and 3% of mean square error (MSE).

show abstract

Earthquake Prediction for Turkey: Ensemble Learning Approach

Tekin,

Sari

2024

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Deep Multi-scale Dilated Convolution Neural Network with Attention Mechanism: A Novel Method for Earthquake Magnitude Classification

Cited by 4 publications

References 24 publications

A class alignment method based on graph convolution neural network for bearing fault diagnosis in presence of missing data and changing working conditions

A class alignment method based on graph convolution neural network for bearing fault diagnosis in presence of missing data and changing working conditions

LUTanh Activation Function to Optimize BI-LSTM in Earthquake Forecasting

Earthquake Prediction for Turkey: Ensemble Learning Approach

Contact Info

Product

Resources

About