This project builds on research in which geo-tagged social network site images are used to determine insect and flowering plant distributions on a continent-wide scale. This work was awarded an "AI for Earth" grant by Microsoft, one of only 6 projects in Australia to receive this recognition.
The new project will build original frameworks for future applications of Machine Learning and Computer Vision techniques to interpret and filter ecologically relevant information from social network site data. The applications and value of such data for scientific research are currently untapped, but potentially enormous. Here, the application area will be invasive species monitoring - a massive problem globally, and one of key importance in Australia.
Invasive species are plants and animals introduced to an area by humans, that cause negative impacts on local ecosystems. This project will investigate spatio-temporal properties of the invasions by pests such as some wasps and ants, cane toads, and flowering weeds. The latest advanced techniques in machine learning and computer vision for image content analysis will be applied to generate data for dynamic species distribution models. This data will in turn be used to generate parameters for simulations of invasive animals and plants.
The use of social network site data to assist in invasive species monitoring will be investigated in this project, a new key to help unlock solutions to managing the problem of invasive species on large, sparsely populated continents like Australia.
- Computer science - programming in C++, Python or equivalent languages.
- Machine learning, neural networks, classifiers etc. an advantage.
- Interest (but not necessarily tertiary qualifications) in ecology an advantage.