5th International Symposium on Machine Learning & Big Data in Geoscience (5ISMLG)
10-13 May 2026, Hong Kong
Data-driven Geotechnical Site Characterization
Speaker:
Jinsong Huang, The University of Newcastle
Abstract:
Geotechnical design and analysis require engineers to make decisions and judgments about ground conditions under conditions of significant uncertainty and incomplete information. It is estimated that, for a typical project, less than one part in a million of the ground volume is directly investigated. Consequently, information obtained from a limited number of investigation locations must be interpolated and extrapolated to represent a much larger subsurface domain. This keynote will introduce data-driven approaches to geotechnical site characterization, ranging from methods based on random field theory and Bayesian updating to more recent machine-learning techniques. Particular emphasis will be placed on tree-based methods incorporating recently proposed geotechnical distance fields. Compared with conventional tree-based approaches, these geotechnical distance fields incur minimal additional computational cost while delivering substantial improvements in performance for both spatial interpolation and soil stratification problems.
Biography:
Jinsong Huang is a Professor at the Priority Research Centre for Geotechnical Science and Engineering within the Discipline of Civil, Surveying and Environmental Engineering at the University of Newcastle. His research focuses on risk assessment in geotechnical engineering and computational geomechanics. He has published more than 250 papers covering topics including slope stability and landslide risk assessment, modelling of spatial variability, stress-integration techniques for elastoplastic constitutive models, contact dynamics of granular media, hydraulic fracturing analysis, and predictive maintenance of railway infrastructure. Professor Huang has received several prestigious awards, including the Regional Contribution Award from the International Association for Computer Methods and Advances in Geomechanics (IACMAG) at its 2014 international conference in Kyoto, the GEOSNet Award from the Geotechnical Safety Network in 2017, and recognition as Australia’s Field Leader in Environmental and Geological Engineering by The Australian 2020 Research Magazine. He has been listed among Stanford University’s “World’s Top 2% Scientists” for single-year impact since 2020 and for career-long impact since 2024. He currently serves as Managing Editor of Georisk and as an editorial board member for the Canadian Geotechnical Journal and Computers and Geotechnics.