‘Big data’ harnessed in grower profitability drive

Author: | Date: 11 Jun 2019

image of tom giles
Tom Giles, GRDC enabling technologies senior manager, says machine learning has the potential to deliver value to Australian grain growers through being used to address a broad range of priority issues Photo: Brad Collis / © GRDC

The Grains Research and Development Corporation (GRDC) is venturing into the transformative field of machine learning to help drive profitability gains for Australian grain growers.

It is investing for the first time in dedicated research projects that address grower priority issues primarily through machine learning – a promising ‘big data’ analytical method that extracts insights from large, unstructured datasets.

Machine learning is the automated creation of algorithms for pattern recognition, classification and prediction from data.

Tom Giles, GRDC enabling technologies senior manager, says machine learning can be a powerful way to analyse data for the grains sector.

“It can help tackle previously intractable problems such as providing accurate establishment, growth stage, head count or crop health estimates for broadacre crops; understanding when and how much nitrogen fertiliser to apply; estimating in-crop frost damage in real time; and scanning publications to extract relevant information to growers in real time,” he said.

“The GRDC has identified machine learning as a foundational technology with the potential to deliver value to Australian grain growers through helping to address a broad range of productions constraints and opportunities outlined in its 2018-23 Research, Development and Extension Plan

“This pilot investment program aims to improve grower outcomes by engaging the machine learning community to develop solutions to specific priority issues.”

Mr Giles said that through the program, the GRDC would help to build Australian grains research capacity in machine learning by supporting the post-doctoral fellowships to be associated with the projects.

“The GRDC will also develop a technical consultation group to facilitate domestic and international collaboration, coordinate efforts and ensure quality control, and provide advice for other investment projects,” he said.

Mr Giles said the GRDC recognised that a number of industry stakeholders were already using machine learning or possessed data sets that could be incorporated into machine learning projects.

“These stakeholders may include public research organisations, state agricultural departments, universities, private pre-breeding and breeding companies, and agritech, agribusiness and agri-input organisations,” he said.

“The GRDC’s machine learning tender therefore includes an open call procurement inviting prospective partners to propose machine learning projects for co-investment.”

Applications are sought for the following investments, in addition to membership of the technical consultation group:

  • Identifying genetic contributors to crop stress tolerance in the presence of environmental effects
  • Delivering accurate and localised weather forecasts
  • Extracting value from crop/soil variability mapping
  • Extracting value from agronomy and farming systems R&D datasets
  • Delivering a Natural Language Programming-driven Question-Answering research repository
  • Open call for Machine Learning Use-Cases delivering value for Australian grain growers.

More information about the ‘Machine learning, capacity, capability and use-cases for the grains industry’ tender is available on the GRDC website. Tenders close on July 2.

The GRDC is a global leader in grains industry research and development. Its focus is on driving discoveries and innovations that will create enduring profitability for Australian grain growers.

Contact Details

For Interviews

Tom Giles, GRDC enabling technologies senior manager,
0417 889 860
tom.giles@grdc.com.au

Contact

Natalie Lee, GRDC communications manager – west
0427 189 827
natalie.lee@grdc.com.au

GRDC Project code: PROC-9175954, PROC-9175955, PROC-9175956, PROC-9175957, PROC-9175958, PROC-9175959, PROC-9175960