Determining shear wave velocity using artificial neural networks

Published: 31 July 2023| Version 1 | DOI: 10.17632/nnt7kwdr39.1
Kwaku Duah


A project to determine shear wave velocity using artificial neural networks. Python was used to build the model and 80 percent of the dataset used to train the data. Multiple Regression, Keras Deep Learning framework was used to model the network. Sklearn with MLPRegressor was also used to build the model. In end, the Multiple Linear Regression returned with a coefficient of determinaion of 0.44 and the Keras Deep Learning framework yielded a coefficient of determination of 0.39 along the MLPRegressor model Note: The data file used to build this project has been uploaded, to try the model, please direct message me, for the password to decrypt the file.


Steps to reproduce

Python was used to manipulate the datasets with scikit-learn in Jupyter Notebook. Data cleaning, data preparation and choice of algorithm are all in this link.


Kwame Nkrumah University of Science and Technology


Artificial Intelligence, Neural Network, Adversarial Machine Learning