Dataset in support of the paper 'Recovering shear stiffness degradation curves from classification data with a neural network approach'
DOI: 10.5258/SOTON/D2101

This work forms part of the activities of the Centre of Excellence for Intelligent & Resilient Ocean Engineering (https://www.southampton.ac.uk/iroe) and the first and third authors are supported by Royal Academy of Engineering under the Chairs in Emerging Technologies scheme. 

Charles J A, University of Southampton j.a.charles@soton.ac.uk
Oztoprak S, Istanbul University
Gouvenec S M, University of Southampton

Journal:Acta Geotechnica
Title:Recovering shear stiffness degradation curves from classification data with a neural network approach
Article DOI:https://doi.org/10.1007/s11440-023-01879-4

Data is released under a CC0 license.

This dataset consists of digitised shear stiffness degradation data taken from 23 publications along with eight associated soil classification parameters. 
The data was originally produced by a range of laboratory and in-situ testing. 
A total of 240 stiffness degradation curves and sets of soil classification parameters are included. 
This dataset was extracted from, edited, and reformatted from an earlier dataset collated by Sadik Oztoprak which was analysed in Oztoprak and Bolton (2013) and has been shared with all necessary permissions. 
This new dataset has been used for work on recovering shear stiffness degradation curves using Artificial Neural Networks.

The dataset is stored in a Matlab file titled StiffnessCurvesAndParameters.mat. 
The .mat file contains the following Matlab cell objects:
references - This cell object has 23 elements, each containing a string naming the a source publication.

properties - This cell object has 23 elements corresponding to the aforementioned list of references. 
Each element contains a cell object which contains an array of soil classification parameters for each test associated with that particular reference.

stiffnessCurves - his cell object has 23 elements corresponding to the aforementioned list of references. 
Each element contains a cell object which contains an array of strain and stiffness values for that particular test.

propLabels - This cell object contains strings for the names of the eight soil classification parameters included in the properties cell object. 

The following parameters are included in this dataset:
'Mean Effective Stress (p)'
'Mean Effective Stress/Reference Atmospheric Stress (p/pa)'
'Overconsolidation Ratio (OCR)'
'Void Ratio (e)'
'Relative Density (Dr)'
'Average Grain Size (D50)'
'Uniformity Coefficient (Cu)'
'Initial Elastic Shear Modulus (G0)'

The digitised data from the following publications is included in this dataset:
'Iwasaki et al 1978'
'Iwasaki et al 1978'
'Tika et al 2003'
'Kokusho and Esashi 1981'
'Hardin and Kalinski 2005'
'Yamashita and Suzuki 1999'
'Fioravante et al 1998'
'Lo Presti et al 1997'
'Yamashita and Toki 1995a'
'Yamashita and Toki 1995b'
'Porovic and Jardine 1995'
'Dong et al 1995'
'Park 1993'
'Yang 2007'
'Yasuda et al 1996'
'Alarco-Guzman et al 1989'
'Seed et al 1986'
'Rollins et al 1998'
'Yasuda and Matsumoto 1994'
'Yasuda and Matsumoto 1993'
'Saxena et al 1988'
'Arango 2006'
'Maher et al 1994'