Our research

PRaDA's innovative research projects are finding solutions to problems in fields such as health care, social media, advanced manufacturing and surveillance. 

Our research has resulted in nine patents. These patents have led to start-up companies including the award-winning iCetana, a video surveillance platform that uses analytics to detect potential security threats in large data sets. The system is now used around the world.


Our research focus

At PRaDA our methods are grounded in statistical machine learning and pattern recognition. We are focused on probabilistic models and optimisation and deep learning.

Probabilistic model research

Our work includes probabilistic graphical models, such as Markov models, hierarchical hidden Markov models, conditional random fields, and more recently Bayesian non-parametric and Bayesian optimisation models.

We are researching what models will be computationally scalable when tackling big data problems. We want to find out:

  • how we can enhance the expressiveness of our models
  • what the representations are 
  • whether we can develop appropriate inference and learning. 

Compressed sensing and sparsity modelling

Our research in this area asks what the sparse, robust models we can construct for big data are. We're also interested in examining lean data produced in the experimental processes to design products efficiently.

PRaDA projects in action

Our innovative projects have a range of applications for the real world. We can use data insights to help doctors predict suicide risks, children with autism learn new skills and neighbourhoods reduce crime.

Assistive Technology and Autism Research

Our research into early intervention therapy has produced a learning tool for children living with autism.

Find out more about the positive therapy results

Health analytics

Our health analytics program conducts research in dynamic health intervention and improved health care and safety.

Find out more about our health start-up and autism app

Large-scale surveillance

Our research into the issues surrounding large-scale surveillance has enabled operators to focus on the crucial 1% of events in a video feed.

Find out more about our breakthrough surveillance algorithm

Social media analysis

Our work in social media analysis focuses on the study of patterns in social media to extract socially meaningful indices that can be used to derive understanding of communities and individuals.

Find out more about our work analysing social media

Contact us

Centre for Pattern Recognition and Data Analytics (PRaDA)
General enquiries
+61 3 5227 1266
Email PRaDA

School of Information Technology
Deakin University
Locked Bag 20000
Geelong VIC 3220