A team from Queensland University of Technology has developed a deep learning tool to streamline the repatriation of Indigenous ancestral remains. Funded by the Australian Research Council and in partnership with various institutions and the Research, Reconcile, Renew Network, this tool tackles the challenge of sifting through historical records from 1790 to 1970. The study addresses the limitations of museum catalogues and the difficulty of analyzing non-digital historical documents. By employing an Informed Neural Network trained with pre-specified matter specific keywords and a small amount of labelled data, the researchers created a model that efficiently locates pertinent documents. This advancement has significantly contributed to the global initiative of returning Indigenous human remains, furthering the efforts of the Research, Reconcile, Renew Network’s mission initiated in 2014.
QUT has published a news coverage on this work at the following link.
Link: https://research.qut.edu.au/qutcds/2024/02/16/machine-learning-indigenous-remains/
The National Indigenous Times has published a news article on this work at the following link.
Link: https://nit.com.au/16-02-2024/9828/data-science-shown-to-expedite-return-of-ancestral-indigenous-remains