Novel approaches to tackling complex data
More Information
Recording below
As the volume of data collected for research continues to grow at a rapid pace, understanding and dealing with its complexity becomes crucial. Complexity looks different across disciplines and can emerge from working with big data, using new techniques and technologies, or combining multiple data sources.
Panelists:
Dr Vanessa Ferdinand
Research Fellow, Computational Cognitive Science, Melbourne School of Psychological Sciences
Vanessa Ferdinand works on computational approaches for understanding cognition, culture, and sociality in humans and animals. She received her PhD in the Evolution of Language from the University of Edinburgh and then took up a prestigious Omidyar Fellowship in Complex Systems Science at the Santa Fe Institute. Currently, she is a Research Fellow at the University of Melbourne’s Complex Human Data Hub. She met her favourite complex data set in 2018, when she volunteered as a field assistant for the Tambopata Macaw Project in the Peruvian Amazon rainforest and is now working as an analyst on that project.
Dr Pip Karoly
Senior Research Fellow, Biomedical Engineering, Data Scientist, Seer Medical
Philippa (Pip) Karoly is working to develop an innovative, patient- specific approach to seizure forecasting. Using sophisticated computational techniques, long-term data from brain recordings, environmental, behavioural and physiological factors can be combined and converted into useful seizure likelihood models.
Zeb Nicholls
Research Fellow Emissions Pathway, Geography, Earth and Atmospheric Sciences
Zebedee is an expert in reduced complexity climate modelling. His research focuses on the development, evaluation and application of reduced complexity models with a particular focus on the Model for the Assessment of Greenhouse gas Induced Climate Change (MAGICC). In the IPCC’s Sixth Assessment Report, he led the writing of Cross-Chapter Box 7.1 on reduced complexity models used for scenario classification in AR6, was a Contributing Author to WG1 Chapters 1, 4, 5, 6, 7 and Technical Summary and WG3’s Summary for Policy Makers, Chapter 3 and Annex C. He can be found on GitHub and GitLab, username znicholls.
Professor Michael Kirley
School of Computing and Information Systems and Co-Director, Melbourne Centre for Data Science
Michael’s research interests encapsulate artificial intelligence, machine learning techniques and game theory, with a strong focus on evolutionary computation. He has made significant contributions to research focused on the design of data-driven algorithms for optimisation and decision-making. Beyond established boundaries and discipline norms, Michael’s work bridges data science and social science to identify solutions to complex social-technological-ecological problems.
Questions that may be explored include:
- When and how does data become complex?
- How are emerging technologies redefining data complexity?
- How have research methodologies adapted to deal with complex data?
- How does complexity change as the volume of data increases?
- How do we choose what data to collect, analyse and inform decisions?
The panel was followed by open discussion and Q&A.
This was a hybrid event.
Hosted by Melbourne Data Analytics Platform (MDAP) and the Melbourne Centre for Data Science (MCDS).