Artificial Intelligence and Data Analytics Research Cluster

The Artificial Intelligence and Data Analytics (AIDA) Research Cluster carries out and promotes top-quality research in data science, machine learning, optimisation and artificial intelligence.

Major research areas

We cover a broad spectrum of research themes with the goal of making artificial intelligence (AI) an integral part of our daily lives.

We develop fundamental AI techniques and apply them to business and commerce, bioinformatics, e-health, sport, defence, environment, precision agriculture, energy, logistics and multimedia.

Artificial intelligence

Our research focuses on aggregation operators in AI, fuzzy systems and image processing. Fuzzy systems employ the concept of partial truth to mimic human decision-making. We build computer systems that help people make decisions from fragmented evidence and explain the logic behind those decisions.

We also focus on processing images, videos and other data to help develop autonomous systems that rely on image sensors, such as robots, rovers and drones.

Data analytics

Data analytics is about discovering patterns and properties of big and small data and developing data-driven models. Our research focuses on signal processing, smart sensing, anomaly detection, and aspects of information security and privacy.

Algorithms and optimisation

Our research covers industrial models and solutions based on discrete and continuous numerical optimisation, global optimisation, constraint satisfaction, models on graphs and data analysis methods.

We also cover development of powerful and efficient algorithms, including parallel programming and graphic processing units. We provide fundamental techniques at the core of many AI algorithms, including machine learning, clustering and regression.

Major research projects

Novel audio watermarking techniques for tracing multimedia piracy

ARC Linkage Projects; Industry partner: Flag Explore

With the rapid growth of communication networks and the use of advanced multimedia technology, digital multimedia data can be easily copied, manipulated and distributed. This has led to strong demand for new techniques to prevent illegal use of copyrighted data.

In this project, Deakin University and Flag Explore will develop inaudible, robust and high-capacity audio watermarking techniques to trace the illegal copying and distribution of multimedia data containing a sound component, such as audios and sound videos.

The outcomes of the project will greatly advance audio watermarking research, strengthen Australia’s competitiveness in this research area, and help prevent huge financial and job losses in the Australian multimedia industry.

An optimisation-based framework for non-classical Chebyshev approximation

ARC Discovery grant

This project aims to solve open problems in multivariate and piecewise polynomial approximation – two directions that correspond to fundamental obstacles to extending classical approximation results.

Through an innovative combination of optimisation and algebraic technique, this project intends to develop foundations for new results in approximation theory and new insights into other areas of mathematics (notably optimisation).

The techniques and methods developed should also have significant benefits in the many disciplines where approximation problems appear, such as engineering, physics and data mining.

Mining patterns and changes of wave shapes for efficiently querying periodic data streams

ARC DECRA grant

Data is becoming the world’s new natural resource, and big data use grows quickly. Useful patterns are often buried in big stream data. In this project, we re-represent raw big stream data using only critical data (or useful information) to reduce the data size and capture the principal features of the patterns to further simplify the internal data correlation.

In addition to achieving efficiency, we also simplify the data analytics problem since it's much easier to discover useful knowledge from very few simple-structured pattern-relevant data. For example, we've demonstrated that data streams can be reduced to less than 2% of their raw size without losing important information, including reduction of medical sensor stream data (to 0.85% of original size), GPS trajectory data (1.7%) and short text stream data (1.25%).

Decision-making systems for developing and improving online games

Funded by Beijing Shandesitong Technology

In this project, we aim to develop offline game decision-making systems to uncover and understand useful patterns underlying game data. These systems can be used to guide the game companies to improve their existing online games and develop new online games.

The key part of this project will be the design and implementation of generic data analysis and processing models. Built upon the models, the profitable potentials for game improvement and development will be investigated.

Scheduling and timetabling for pilot training

Funded by DSTG ALGORA

This project aims to develop a highly intelligent planning tool that integrates simulation, optimisation and data analysis for the daily scheduling of training lessons while, at the same time, allocate limited resources under complex resource restrictions.

Research funding

Category 1: Australian competitive grants

  • Novel audio watermarking techniques for tracing multimedia piracy
    Research team: Yong Xiang, Wanlei Zhou, Gleb Beliakov and Longxiang Gao
    Funding: ARC Linkage Projects (LP170100458), Australian Research Council, Australia, 2018–2021
  • An optimisation-based framework for non-classical Chebyshev approximation
    Research team: Julien Ugon, Marco Lopez-Cerdá, Nadeza Sukhorukova, Vera Roshchina, Jean-Pierre Crouzeix and Nira Dyn
    Funding: ARC Discover Projects (DP180100602), Australian Research Council, Australia, 2018–2020
  • Mining patterns and changes of wave shapes for efficiently querying periodic data streams
    Researcher: Guangyan Huang
    Funding: ARC DECRA Project (DE140100387), Australian Research Council, Australia, 2014–2016
  • Deep data mining for anomaly prediction from sensor data streams
    Research team: Yu Zhang and Guangyan Huang
    Funding: ARC Discovery Projects (DP140100841), Australian Research Council, Australia, 2014–2016
  • Developing an environmentally friendly, low cost solution to reduce wear and improve productivity in metal forming
    Research team: Bernard Rolfe, Peter Hodgson, Maria Forsyth, Yong Xiang, Matthew Doolan and Micheal Pereira
    Funding: ARC Linkage Projects (LP120100239), Australian Research Council, Australia, 2012–2014
  • Blind separation of mutually correlated sources
    Research team: Yong Xiang and Yue Rong
    Funding: ARC Discovery Projects (DP110102076), Australian Research Council, Australia, 2011–2013
  • Tracing real Internet attackers through information correlation
    Research team: Yang Xiang, Wanlei Zhou and Yong Xiang
    Funding: ARC Discovery Projects (DP1095498), Australian Research Council, Australia, 2010–2012
  • Capturing dance: using motion capture to enhance the creation of innovative Australian dance
    Research team: Kim Vincs, Vicki Mak and Richard Smith
    Funding: 2009–2011
  • Blind signal separation from unidentifiable systems
    Research team: Yong Xiang and Hieu Trinh
    Funding: ARC Discovery Projects (DP0773446), Australian Research Council, Australia, 2007–2009

Category 2-4: Other public sector, industry and cooperative research centre grants

  • Decision-making systems for developing and improving online games
    Research team: Yong Xiang and Longxiang Gao
    Funding: Beijing Shandesitong Technology, China, 2018–2019
  • Heartarc – Improving heart health and wellbeing
    Research team: Kon Mouzakis, Rajesh Vasa, Mohamed Abdelrazek, Andrew Cain, Gery Karantzas and Thanh Nguyen
    Funding: 2017–2018
  • Timetabling solutions for defence
    Research team: Vicki Mak and Sergey Polyakovskiy
    Funding: DST Group, Australia, 2017–2018
  • Information oriented fog servers in national parks
    Research team: Yong Xiang, Longxiang Gao and Atul Sajjanhar
    Funding: Telematics Trust, Australia, 2017–2018
  • Developing a location aware fog computing prototype used on vehicles
    Research team: Longxiang Gao and Yong Xiang
    Funding: Flag Explore, Australia, 2017–2018
  • Communicating Victoria’s environmental condition: next steps for the Yarra and Bay website and its report card analysis tool INT274
    Research team: Research team: Yong Xiang and Peter Dahlhaus
    Funding: Environmental Protection Authority Victoria, Australia, 2017
  • Developing a preliminary image pre-processing algorithm to identify images containing wild animals
    Research team: Yong Xiang and Longxiang Gao
    Funding: Parks Victoria, Australia, 2016
  • Improving wireless server capacity and efficiency using advanced wireless techniques
    Research team: Yong Xiang and Longxiang Gao
    Funding: Flag Explore, Australia, 2016
  • Development of database system for descriptive and statistical information storage and analysis
    Research team: Yong Xiang and Longxiang Gao
    Funding: Parks Victoria, Australia, 2015
  • Development of advanced wireless monitoring system for emergency lighting
    Research team: Abbas Kouzani and Yong Xiang
    Funding: Research Connections Program, Department of Industry, Australia, 2015
  • Miniaturisation and optimisation of a wireless smart remote monitoring system for emergency lighting
    Research team: Abbas Kouzani and Yong Xiang
    Funding: MPower Group, Australia, 2014
  • Optimisation of a wireless electronic device for longer battery life and lower energy consumption
    Research team: Yong Xiang and Abbas Kouzani
    Funding: Planet Innovation, Australia, 2013
  • Development of a smart emergency lighting management system
    Research team: Yong Xiang and Abbas Kouzani
    Funding: Researchers in Business Scheme, Australian Institute for Commercialisation, Australia, 2010–2011

Our staff

Group directors

NameExpertise
Professor Yong Xiang Data analytics, machine intelligence, signal and image processing
Professor Gleb Beliakov Fuzzy systems, computational intelligence, numerical approximation and machine learning

Group members

NameExpertise
Professor Maia Angelova Data analytics, e-health
Professor Lynn Batten
Professor Antonio Robles-Kelly AI and machine learning
Associate Professor Richard Dazeley Multi-objective reinforcement learning
Associate Professor Tim Wilkin

AI, fuzzy systems, aggregation operators

Dr Shaun Bangay

AI and virtual reality

Dr Longxiang Gao Data analytics, Fog computing, Information privacy
Dr Jingyu Hou Data analytics, data and web mining, bioinformatics algorithm design, databases and information retrieval
Dr Guangyan Huang Data mining, parallel algorithms for big data, image/video analysis
Dr Simon James Fuzzy logic
Dr Wei Luo

Data analytics, machine learning

Dr Vicky Mak-Hau

Optimisation, exact algorithms, meta-heuristics

Dr Kerri Morgan

Graphs model networks, biological networks, transport networks, and social networks

Dr Asef Nazari

Optimisation

Dr Thanh Nguyen

Computer vision and machine learning

Dr Sergey Polyakovskiy

Optimisation, constraint programming

Dr Atul Sajjanhar

Image processing, face recognition

Dr David Tay

Wavelet analysis, image processing, computational data analytics, machine learning

Dr Julien Ugon

Nonsmooth optimisation, optimisation algorithms, approximation theory

Dr Leo (Yu) Zhang

Compressed sensing, cloud and multimedia security

Contact us

Director
Professor Yong Xiang
+61 3 9251 7740
Email Prof. Xiang

Director
Professor Gleb Beliakov
+61 3 9251 7475
Email Prof. Beliakov