Networks have been extensively used to capture social interactions, by representing individuals as nodes and their relationships as edges.
Such networks have been used to model the spread of epidemics. A few nodes are 'infected', and over time they gradually infect their neighbours on the network, who in turn infect their neighbours, etc. This type of model can then be used to simulate different intervention strategies aimed at containing outbreaks.
However, an important limitation is the difficulty to visualise these networks when they approach a realistic scale.
For instance, imagine we are interested in a specific city (Brisbane), a given disease (COVID-19), and that we have a network-based model to study the outbreak. How would we visualise the spread of the disease over a network of 2.2 million nodes? On many computers, we would only have one pixel per node.
This project will require you to apply your computer science skills in order to design and implement a hierarchical visualisation method that allows users to view information at the level of the entire network but also to zoom into specific areas for a more detailed view.
A starting point could be to consider methods that have been proposed to analyse networks, such as community detection methods.
In this project, you will review literature and existing network analysis tools in order to gain an understanding of what is currently available.
You will have the opportunity to write your own code to visualise large networks.
During this, you will gain experience in visualisation, network analysis and epidemic modelling, and have the opportunity to work alongside researchers working in modelling and data science.
Skills and experience
This project requires you to have strong programming skills. Some experience with Python or R (and their respective visualisation libraries) would be beneficial.
No prior experience with network or epidemics is necessary, but you must be willing to learn relevant concepts.
You may be eligible to apply for a research scholarship.
Contact the supervisor for more information.