A consilience partnership for new data science and new discoveries in science, health and society


Identifying and understanding the societal significance of premodern artefacts

The project has two distinct aims:

i) To create data repositories of presently undigitized pre-modern artifacts held by the Art Gallery of NSW. The design of these data repositories will be informed by the scholarship of data collection and curation, including identifying potential cultural sensitivities and understanding of patterns of collection and acquisition; 

ii) To develop new data science tools that will enable historical and archaeological data sets to be intelligently analysed, both at an individual item and aggregated narrative level, to support discoveries across the domains of history and archaeology, supported through the FAIME fieldwork archaeology framework.

The project forms a planned collaboration between DataX; Google Arts and Culture; and the Art Gallery of NSW (both with whom Macquarie has an existing relationship). For DataX, the value lies in developing new and innovative data science tools that can develop narratives across data and be used in sectors of history and archaeology to generate new hypotheses and discoveries.

Lead CIs: Prof Malcolm Choat, Dr Susan Lupack, Prof Louise D’Arcens, Prof Niloufer Selvadurai, Dr Richard Savery and Prof Mark Dras.

Brain networks for health

Poor understanding of speech in noise continues to be a perplexing problem in hearing science. Approximately one in ten people with speech-in-noise hearing difficulties cannot be helped because the underlying cause of their hearing problem remains a mystery.  It is proposed that these types of hearing problems relate to dysfunction in a brain network that separates signal (speech) from noise. Understanding this network and how it might contribute to restoring the recognition of speech in noise is a problem of significant importance. 

DataX will develop a new data-centric approach to brain network analysis: connectome-based predictive modelling, to predict individual speech in noise performance from brain connectivity. We will uncover the most predictive features (i.e., brain connections) that enable speech in noise perception. The analysis techniques developed will lay foundations for data mining of multiple sources of brain data as they pertain to other aspects as-yet unsolved neurological challenges. Ultimately, the structure of the human connectome could be used to inform artificial agents that are able to mimic the way a human interacts with the world, both in health and disease. 

Bringing cosmic fossils to life: Uncovering the history of galaxy formation with MUSE

The answers to solving the origins of complexity in our Universe lie in combining theory and observation in new ways, requiring an integrated approach to data and modelling. DataX will bring together computer scientists, statisticians and astronomers to develop new ways of bringing theories, mathematical models and data together to generate new understandings and explanations of the complexity of galaxy formation. Complexity takes many forms in the Universe around us, from the extremes of size, mass and energy on cosmic scales; to the diversity of life on our own planet. Yet the origin of the Universe is one of surprising simplicity, starting at the so-called Big Bang with a near-uniform plasma composed of mostly hydrogen and a handful of other elements. How did the complexity we see today build over the 13.8 billion years since the Big Bang? In the coming decade, astronomers will be exposed to a new level of detailed observations about how the Universe works, thanks to powerful new telescopes and instruments becoming available, some of which are currently being developed at Macquarie University. At the same time, increasingly complex computer modelling is required to capture and test our understanding of the physical processes at play. This will be a challenging project developing new mathematical data science and computing approaches to enable new discoveries in science.

Superannuation and financial products

Australia’s life insurance, superannuation and pension funds industries carry significant responsibility for the financial wellbeing of Australians. Managing this responsibility and financial risk depends on accurately pricing consumers’ insurance premiums. To set those premiums, the industry analyses data to make predictions about individual mortality, yet technological advances have produced unprecedented volumes and sources of possible data to choose from and merge. This makes life expectancy forecasting and premium-setting potentially inaccurate. This project will develop new theories, methodologies and algorithms that account for complexities in merged big datasets to improve the accuracy of predictions. Translated into a purpose-built open access software program coupled with industry practitioner training, our research will build industry’s capacity to use these new methodologies leading to improvements in mortality forecasts and pricing of life insurance premiums for everyday Australians, as well as stronger financial risk management among some of Australia’s most critical financial industries.