This transdisciplinary project helps public decision-makers tackle complex public problems by using AI in new and innovative ways. We are undertaking social and policy research to understand the needs and concerns of decision-makers to ensure the AI solutions we develop address these, while at the same time building AI tools that analyse data and generate actionable insights in ways that empower decision-makers to make evidence-informed decisions and support the achievement of sustainable development.
Data science pathways for sustainable livelihoods in remote indigenous Australia
This project establishes a platform to achieve Traditional Owner led data science research that integrates traditional Indigenous knowledge with empirical interdisciplinary science.The purpose is to harness the potential of data science to co-design a community-owned platform that supports self-determined sustainable livelihoods in Australia's most remote Indigenous homeland communities. This research is a partnership between the Centre for Appropriate Technology, Olkola Aboriginal Corporation, Yintjingga Aboriginal Corporation, Monash University and The University of Melbourne.
Rory Taylor, Monash Sustainable Development Institute
AI and Machine Learning for predicting species responses to global change
This project brings together advances in biological modelling, machine learning, and artificial intelligence to determine whether alternative low-cost data sources combined with new multi-paradigm modelling techniques can transform our capacity to predict and manage biological responses to global change.
Prof Melodie McGeoch, School of Life Sciences, La Trobe University
Predicting flood risk: The case of the Citarum River,Indonesia
Flooding has devastating effects, bringing loss of life, homes, property, mobility and livelihoods and in developing cities brings disease, pests, food and clean water shortages. Flooding is an escalating challenge due to urbanisation and climate change. Traditional approaches for flood prediction use high quality flood data as inputs to complex hydrodynamic models. Developing countries without quality data need new tools to support disaster planning and responses. Our project is exploring the technical feasibility of using satellite data and machine learning to predict flooding of the Citarum River in West Java, Indonesia, one of the planet's most heavily populated and polluted rivers.
Dr Dwinanti Marthanty, Civil Engineering, Universitas Indonesia
Combining rapid evidence synthesis and expertise for threatened ecosystem recovery
This project is developing a novel approach to supporting evidence-based decision-making in conservation management. Drawing on tools for rapid evidence synthesis and expert consultation in the health sciences, researchers are creating a standardised approach that can efficiently identify effective management actions to address important conservation problems. Initial case studies are focused on threatened woodlands and peatlands in Australia.
Net zero precincts: Citizen data commons and technological sovereignty
Local sustainability initiatives often lack meaningful community engagement in data governance. This project developed a novel participatory approach to enable multiple stakeholders to co-design and co-appraise data governance prototypes in the context of the Monash Net Zero Precinct. Results reveal the importance of harnessing community engagement to reflect the contexts, values and interests of diverse stakeholders and empower multilevel participation in data governance.
Prof Lyn Bartram, School of Interactive Arts + Technology, Simon Fraser University
Towards SDG 4: Transferring forum brain's success to health sciences and beyond
Over 100 million people learn on a Massive Open Online Course (MOOC), a powerful force for achieving widespread educational goals in SDG 4, and in every single course, an online discussion forum is used. Yet, no automated human-quality grading and feedback technology for open ended forums exists. ForumBrain AI, built by SoDa Laboratories in the Monash Business School, aims to be the solution to this problem, leveraging state of the art Natural Language Processing to make powerful and safe technology available to teachers and students alike.