eCite Digital Repository

A speedy update on machine learning applied to bedrock mapping using geochemistry or geophysics: examples from the Pacific Rim (and nearby)

Citation

Hood, SB and Gazley, MF and Cracknell, MJ and Barker, S and Reading, AM, A speedy update on machine learning applied to bedrock mapping using geochemistry or geophysics: examples from the Pacific Rim (and nearby), Proceedings of the 2019 Mineral Systems of the Pacific Rim Congress (PACRIM 2019), 3-5 April 2019, Auckland, New Zealand, pp. 58-61. (2019) [Refereed Conference Paper]


Preview
PDF
915Kb
  

Copyright Statement

Copyright unknown

Official URL: http://pacrim.ausimm.com/

Abstract

Artificial Intelligence (AI) has numerous and varied definitions, leading to confusion and disagreement about what it represents, and how it relates to mineral exploration. A general definition is most effective: AI is the artificial machine-based reproduction of tasks inspired by, or conventionally accomplished by humans (or other animals) using their natural intelligence. The form of AI applied to mineral exploration is called ‘Domain Specific AI’; it is task-oriented and includes decades-old approaches to software automation and statistical modelling.

A core skill of experienced economic geologists is pattern recognition. This might include field-based work (e.g., recognising groupings of rock types, alteration minerals, or mineral textures related to mineralisation) or laboratory-based work (e.g., identifying groups of similar analyses from geochemical assays, or categorising spectral data from remote sensing devices). Machine learning (ML) is a subfield of AI that specialises in pattern recognition and is defined as any computer program that improves its performance at some task through experience or iteration. ML is well-studied and has routinely been applied towards mineral exploration over five decades. ML can automate parts of mineral exploration workflows, e.g., mapping or modelling geology, and can improve results by making them more objective, repeatable, or efficient.

This extended abstract briefly gives three examples of ML used to improve the interpretation of rock type in a mineral exploration or mining area, using geochemical or geophysical data.

Item Details

Item Type:Refereed Conference Paper
Keywords:machine learning, geological mapping, feature selection
Research Division:Earth Sciences
Research Group:Geology
Research Field:Geology not elsewhere classified
Objective Division:Mineral Resources (excl. Energy Resources)
Objective Group:Mineral Exploration
Objective Field:Mineral Exploration not elsewhere classified
UTAS Author:Hood, SB (Mr Shawn Hood)
UTAS Author:Cracknell, MJ (Dr Matthew Cracknell)
UTAS Author:Barker, S (Dr Shaun Barker)
UTAS Author:Reading, AM (Professor Anya Reading)
ID Code:132180
Year Published:2019
Deposited By:CODES ARC
Deposited On:2019-04-26
Last Modified:2019-06-20
Downloads:0

Repository Staff Only: item control page