Webinar: Geological Knowledge Discovery using Machine Augmented Intelligence
Geological interpretation is a complex task where an interpreter’s bias plays an important role. As a result, interpretation outcomes are variable and uncertain.
About this event
With the increasing use of artificial intelligence and machine learning in our daily lives such as for information search, online shopping, and virtual assistant AIs, the geoscience domain has also been active in the uptake of machine learning and AI to assist in interpreting geology from data.
This talk presents innovative machine-assisted technologies that improve the efficiency and the robustness of geological interpretation of different types of geodata used in the resource industry.
A number of applications of machine learning were developed in collaboration with the mining industry for the analysis and integration of multi-modal drill hole data. These applications integrate the algorithms and workflows to assist human decisions. The approach is to provide end users the control of the algorithmic process as much as possible; and to enable a seamless integration of algorithms in the interpreter’s workflow using interactive visualisation.
This talk also presents ongoing AI research that extracts geological insights from documents using machine reading of text. It applies advanced text mining methods and constructs a graph-based knowledge base called a knowledge graph to store and access geological information. Case studies on different mineral deposits demonstrate the effectiveness of the methods for rapidly and robustly transforming text data into structured information that faithfully represents the contents of the source reports.
About the speaker
Professor Eun-Jung Holden from The University of Western Australia (UWA) has leading expertise in the application of data science in earth sciences. She leads Centre for Data-driven Geoscience (CDG) at UWA and is the Director of UWA Data Institute.
Her research team, CDG, spans the boundaries of computational science and geoscience. The team works closely with the resource industry to develop innovative and deployable data science solutions to semi-automate or automate the process of geological interpretation from diverse types of geodata.
Her team’s research outcomes are disseminated to global end-users through the commercialisation of CET Grid Analysis and CET Porphyry Detection extensions for Seequent’s Oasis Monaj; and televiewer image analysis methods in the Image & Structure Interpretation Workspace for ALT’s WellCAD. The team’s recent research also resulted in three industry driven patent applications on machine learning-based methods for drillhole data interpretation.
She leads a major industry funded research engagement between UWA and Rio Tinto Iron Ore to assist the modelling of geology and material compositions from different types of data including geological, geochemical, geophysical, spectral and photographic data.
Professor Holden’s team won the UWA Vice Chancellor Award in Impact and Innovation in 2015 and she was a winner of the Women in Technology in WA (WiTWA) Tech [+] 20 Awards in 2019. She is a Board member for Advanced Mining Technology Centre at the University of Chile, and for Centre for Business Data Analytics at UWA.
Date and Time
1.00pm – 2.00pm (UTC+08:00)