We are proud to report that a Bachelor of Data Science and Decisions student has joined a global science challenge for students to help improve bushfire management. 

Final year student Ellen Wang has been participating in the challenge which aims to build AI-enabled technology that can streamline and predict the path of bushfires. 

The project, which is being led by NASA, Microsoft, Geoscience Australia, The Country Fire Authority and EY, plans to develop a tool that will save valuable time when battling bushfires, and ultimately save lives and property in emergency situations. 

"I participated in the first phase of the challenge which is, given a series of satellite images taken from the Australian bushfires, classify whether the pixels within the images were on fire or not", Ellen said. "The current method used to identify and keep track of these bushfires is very manual and time-consuming, so to have a technology that helps speed up this process would be immensely beneficial for the community in terms of resources saved, lives saved, livestock, property, and earlier warnings in the case of a bushfire.

"I got involved because I saw it as a great opportunity to give back to the community; anyone located in Australia would know of the devastating impacts of the 2019-2020 bushfires so it’s definitely an issue that’s very close to heart and close to home.

"I also saw it as a chance to learn more about image datasets, which is something I’m not quite used to because as a data scientist I’m typically found analysing and looking at numbers. It’s a great opportunity for anyone interested in learning more about data science, computer vision, or just wanting to give back to the community", Ellen said. 

The challenge has received media coverage. Ellen, who was UNSW DataSoc's 2020 President and is currently undertaking an internship with Swiss Re, was interviewed on ABC News last weekend about her participation in the project, along with EY Data Science Challenge Leader Olivia Sackett.

You can watch the full ABC interview here