Skip to the content | Change text size

Productivity and Innovation - Intelligent Systems

Productivity and Innovation

The theme of Intelligent Systems, with the broader Faculty research priority of IT enabling productivity and innovation is closely aligned with the Faculty’s research centre, the Centre for Research in Intelligent Systems (CRIS).

A diverse range of projects are underway in this area  ranging from: work in immediate retrieval of music based on the actual properties of the music itself; supporting adaptive and interactive documents helping to reduce paper waste; computational creativity; to multi-sensory fusion and understanding in robotic assistive technology environments. Recently concluded projects in this area can be accessed from Intelligent Systems - archived projects.

Our researchers working in this area include the Director of CRIS, Professor Geoff Webb, as well as Professor Ingrid Zukerman, Professor David Green, Professor Guojun Lu, Professor Kim Marriott, Associate Professor Jon McCormack, Associate Professor Bernd Meyer, Associate Professor Maria Garcia de la Banda and Professor Mark Wallace.

Towards realistic verbal interactions between people and computers - A probabilistic approach

Researchers: Prof Ingrid Zukerman
Centre: CRIS
Funding: ARC Discovery Project 2011-2013
Project outline: This project aims to facilitate natural spoken interactions between people and computer systems, addressing obstacles to the acceptance of these systems. more...
We will investigate computational models for relevant aspects of spoken dialogue, which will be implemented in computer systems for diverse tasks (for example, home devices and phone-enabled services). less...
Related theme: Social Inclusion - Designing IS and Information Infrastructure for all: Inclusive, User Sensitive Design
Key outcomes:  

Learning Complex classifiers without search

Researchers: Prof Geoff Webb
Centre: CRIS
Funding: ARC Discovery Project 2011-2013
Project website: www.csse.monash.edu.ua/~webb
Project outline: This project investigates novel approaches to computational data analysis that use new forms of probabilistic models of data. more...
These new approaches complement the state-of-the-art, suiting large quantities of categorical data, being robust in the presence of errors, and efficiently handling updates when new data become available project investigates novel approaches to computational data analysis that use new forms of probabilistic models of data. less...
Related theme/s:

Health and Wellbeing: Bioinformatics/Protein Structure/Others
Productivity and Innovation - Data Management

Key outcomes:  

Modelling and simulation of self-organised behaviour in biological and bio-inspired systems

Researchers: A/Prof Bernd Meyer, A/Prof B D Hughes; A/Prof T Nakagaki
Partners: University of Melbourne, Future University Hakodate
Centre: CRIS
Funding: ARC Discovery Project 2011-2013
Project website: http://www.csse.monash.edu.au/~berndm/LangeSim/
Project outline: Understanding self-organised systems is fundamental in biology and bio-inspired engineering. The project develops sophisticated mathematical modelling techniques and high performance simulation methods for such systems. more...

This will increase our capacity to explain complex biological behaviour and to produce reliable bio-inspired engineering solutions. less...
Related theme: Health and wellbeing - Bioinformatics/Protein Structure/Others
Key outcomes:  

Flexible user-guided network layout for biomedical applications

Researchers: Dr Michael Wybrow
Centre: CRIS
Funding: ARC Discovery Project 2011-2013
Project outline: This project will develop techniques for automatic layout of biological network diagrams, allowing users to guide the layout while satisfying any required placement constraints and drawing conventions. more...
As part of the project, these methods will be integrated into several real-world systems biology applications for network browsing and authoring. less...
Key outcomes:  

System Biology: Modelling and inference of gene regulatory network

Researchers: A/Prof Madhu Chetty, Prof R. Coppel, Prof P. Wangikar & Dr V. Nguyen
Partners: Indian Institute of Technology, Bombay
Centre: CRIS
Funding: Australia-India Strategic Research Fund 2011-2013
Project website: http://personal.gscit.monash.edu.au/~mchetty/
Project outline: In the post-genomic era, holistic understanding of biological systems in all their complexity is critical in comprehending nature’s choreography of life. more...
Biological processes and systems can be abstracted as multi-layered networks interacting with each other to create a complete biological system. Understanding the interactions of genes plays a vital role in the analysis of complex biological systems. The system level view of gene functions provided by gene regulatory networks (GRNs) is of tremendous importance in uncovering the underlying biological process of living organisms, providing new ideas for treating complex diseases, and the designing of new drugs. The on-going research work involves development of novel and sophisticated computer models incorporating time delays, dynamicity, feedback and large scale representation.  We are exploring two approaches: i) Deterministic S-system models based on first order differential equations ii) Probablistic models based on dynamic Bayesian network. less...
Related theme: Health and wellbeing - Bioinformatics/Protein Structure/Others
Key outcomes:  

Proteomics: Drug Discovery Based on 2-Stage Hierarchical Structure Prediction of G Protein Coupled Receptors: From Computations to Pharmaceutics

Researchers: A/Prof Madhu Chetty, Dr David Chalmers, Prof David Green & A/Prof Ram Samudrala
Partners: University of Washington, USA & Faculty of Pharmacy, Monash University
Centre: CRIS
Project website: http://personal.gscit.monash.edu.au/~mchetty/
Project outline: Protein structure prediction from amino acid sequence has been a grand challenge problem in molecular biology. more...
With transmembrane receptor family, namely, G protein coupled receptors (GPCR) being targets for nearly 50% of all currently used therapeutic drugs, an understanding of these receptors and their structures has become vital in drug design. Research on GPCRs has assumed significance as it is involved in the treatment of a plethora of diseases. Briefly, the proposed project aims to generate a set of near optimal candidate folded GPCR proteins (low resolution protein structures) involved in diabetes from the sample conformation space, guided by novel evolutionary search techniques, scoring functions and other sequence-dependent biases. We then fine-tune native-like conformations from these set of near-optimal GPCR protein folds and develop high-resolution tertiary protein structures. We then predict candidate drugs that bind the known target structure with high affinity and selectivity using sophisticated techniques of mathematical and molecular modelling of GPCR/drug interactions. less...
Related theme: Health and wellbeing - Bioinformatics/Protein Structure/Others
Key outcomes:  

Development of an 'ageing household' model for assessing medium to long-term vaccine impact in populations

Researchers: Dr J McVernon, Dr Kevin Korb, Dr K Glass, Dr J McCaw, Dr E McBryde
Partners: University of Melbourne (School of Population Health); ANU (National Centre for Epidemiology and Population Health)
Centre: CRIS
Funding: ARC Discovery Project 2011-2013 (Administered by University of Melbourne)
Project outline: As birth rates in developed and newly industrialising countries fall, so too do the number of households containing children, with implications for the spread of infections in families. more...
We aim to study the influence of this phenomenon on the risk of common childhood infections, and the length of time that vaccines given in infancy will protect. less...
Related theme: Health and wellbeing - Optimal Health: meeting people's needs; optimising resources; minimising costs
Key outcomes:  

Sentiment detection from opinion surveys the quest for customer and employee satisfaction

Researchers: Prof Ingrid Zukerman
Partners: GapBuster Worldwide
Centre: CRIS
Funding: ARC Linkage 2008-2011
Project outline:

Sentiment detection determines whether discourse is subjective, and if so, its polarity, intensity and usefulness. Automatic sentiment detection is essential for discriminating fact from opinion in information gathering applications, and extracting prevalent opinions about items. more...
In this project, we develop computational mechanisms that detect sentiment from texts, and summarize the sentiment expressed in multiple texts.  The developed techniques are applied to textual feedback provided by customers and employees about different aspects of service industries. The implemented computer system will detect the sentiment expressed in this feedback and produce summaries, which will be used to make recommendations for service improvements. less...

Key outcomes:  

Emergence of robust, stable structures via computation within natural networks

Researchers: Prof David Green
Centre: CRIS
Funding: ARC Discovery 2007-2011
Project outline: Nature as computation is a powerful analogy that has proved a rich source of scientific insights and computing methodologies. more...
This project addresses two central problems of natural computing: how self-organisation occurs within connected networks of agents and how global properties emerge from local interactions. These are explored both in living systems, including landscape genetics and social networks, as well as computational systems, especially multi-agent systems. Processes investigated include synchronisation, clustering, enslavement, feedback and phase changes. The results will provide insights into social issues and self-organisation in networks of agents as well as novel methods for solving complex computational problems. less...
Key outcomes:  

Solar biofuel and carbon sequestration using Cyanobacteria

Researchers: A/Prof Madhu Chetty, Prof R. Coppel, Prof P. Wangikar & Dr Vinh Nguyen
Partners: Indian Institute of Technology, Bombay
Centre: CRIS
Funding: Australia-India Strategic Research Fund 2011-2013
Project outline: Bio-fuel production and carbon sequestration have been getting increasing attention in the recent years due to the dwindling fossil fuel reserves and global warming, respectively. more...
Cyanobacteria are oxygen evolving photosynthetic prokaryotes and play a key role in the harvesting of solar energy. Cyanobacteria are also responsible for naturally sequestering a large part of carbon dioxide from the earth’s atmosphere. An emerging idea is to sequester CO2 at source by using cyanobacterial ponds.  Further, in nature, cyanobacteria are exposed to low levels of CO2 and convert most of the carbon to biomass and glycogen, both of which are relatively low value products.  Many of the cyanobacteria also have naturally occurring biosynthetic machinery that is capable of converting CO2 to fine chemicals and fuel.  We propose to develop techniques to investigate, via in-silico and in-vitro experiments, the underlying regulatory processes that channel the carbon toward various products. A system level view provided by reconstructing a genetic network in a genome scale model is crucial in tuning the biosynthetic machinery of these bacteria.  We will focus our attention on four cyanobacteria strains, namely, Synechocystis sp. PCC 6803, Cyanothece sp. ATCC 51142, Cyanothece sp. PCC 7424, and Cyanothece sp. PCC 8801.  We will simulate various genotypic and growth conditions for cyanobacteria using high performance computing platform and involving altered light availability, substrate availability, light-dark cycles, etc while the genotypic conditions may involve knock-out / knock-in of certain regulatory genes. A select subset of these predictions would be experimentally validated via phenotypic characterization. less...
Related theme: Sustainability - Optimisation of Infrastructure
Key outcomes:  

Game Theoretic Approach to Agent-based Modelling for Carbon Trading Market

Researchers: A/Prof Madhu Chetty, Dr Jacob Crandall, Dr Suryani Lim
Centre: CRIS
Funding: Monash University-Masdar Institute of Science and Technology Strategic Fund 2010
Project website: http://personal.gscit.monash.edu.au/~mchetty/
Project outline: Reducing greenhouse gas (GHGs) emission and slowing down the global warming process is one of the greatest contemporary challenges of our time. more...
Carbon trading scheme is being established around the world as an instrument in reducing global GHG emission economically. In this project, we propose to develop novel carbon trading simulator using an agent-based modeling paradigm. The agents’ trading is driven by the concept of Nash equilibrium obtained from game theory. Combining the AB and the game theory approach, the simulator is capable of modelling important aspects of the real world carbon trading market, namely the relationship between carbon credit prices, trading gain or loss, and traders’ profiles (e.g., carbon emissions, Marginal Abatement Cost Curve (MAC), and trading strategies). The simulator is referred as: ACE-GT2 (Agent-based Carbon Emissions – Game Theory Traders). Based on Rich Internet Application (RIA) technology, the ACE-GT2 simulates a web-based game in which the computer agents trade as nation agents under several trading scenarios. In demonstrating the capability of the model and versatility of the simulator software ACE-GT2, a number of experiments with different settings for trading are being designed and simulated. The results are giving insight to two questions that are of great interest to real world carbon traders: (i) the future trend of carbon credit price, and (ii) the relationship between traders’ trading strategies and trading gain. less...
Related theme: Sustainability - Prediction, Preparedness and Response
Key outcomes:  

Integrated Intelligent Decision Support for Field Design and Management of Census Operations in Australia

Researchers: Prof Geoff Webb
Partners: Swinburne University, Australian Bureau of Statistics (ABS)
Centre: CRIS
Funding: ARC Linkage 2007-2011
Project outline: This project contributes to a more reliable and accurate measurement of the number and key characteristics of Australian people by supporting a more efficient and effective field design and management of the Census operations. more...
As the Census provides a snapshot of Australia and is crucial to communities, private institutions and all levels of government in the planning of services and facilities, this project not only addresses the research priority of smart information use, but also contributes to strengthening Australia's social and economic fabric.  This project will also train highly qualified IT specialists critical to Australia's scientific and industrial development, thus increasing our competitiveness in information technology R&D. less...
Key outcomes:  

Computational methods for protein structural comparison and analysis

Researchers: Dr Arun S. Konagurthu, Prof. Arthur M. Lesk (Penn State), Prof. Peter Stuckey (UniMelb), A/Prof Maria Garcia de la Banda, Dr Lloyd Allison , Prof James Whisstock, A/Prof Ashley Buckle
Partners: Pennsylvania State University, University of Melbourne
Centre: CRIS
Funding: Pending
Project website: http://www.csse.monash.edu.au/~karun/Site/research.html
Project outline: Methods to search compare, classify, interpret and investigate protein structure are of crucial interest to understand the principles of protein architecture, function and evolution. more...
Building on our successful previous work in this space, this project intends to develop a suite of methods and software that will provide a comprehensive and powerful structure analysis workbench and resources for structural biologists and crystallographers world over. less...
Related theme: Health and wellbeing - Bioinformatics/Protein Structure/Others
Key outcomes:  

Analysis of biological network data

Researchers: Dr Arun S. Konagurthu, Prof. Arthur Lesk (Penn State), Prof. Peter Stuckey (UniMelb), Dr. Lloyd Allison
Partners: Pennsylvania State University, University of Melbourne
Centre: CRIS
Project website: http://www.csse.monash.edu.au/~karun/Site/research.html
Project outline:

The recent advances in technologies such as Mass Spectrometry, ChIP-chip experiments, Next Generation Sequencing, Yeast Two-Hybrid assays, Combinatorial Reverse Genetic screening and Automated literature mining techniques are fuelling an exponential growth of data whose preferred representation is a graph or a network. more...
This project builds on methods to analyse complex organisational and functional characteristics of various biological networks. less...

Related theme: Health and wellbeing - Bioinformatics/Protein Structure/Others
Key outcomes:
  1. Konagurthu, A.S. & Lesk, A. M. (2008), 'On the origin of distribution patterns of motifs in Biological networks'. BMC Systems Biology, 2:73.
  2. Konagurthu, A.S. & Lesk, A. M. (2008), 'Single and multiple input modules in regulatory networks'. Proteins: Structure, Function, and Bioinformatics, 73(2): 320-324.

Computational methods for analysis of Next-generation genome sequencing data

Researchers: Dr Arun S. Konagurthu, Dr Torsten Seemann (VBC), Dr. Lloyd Allison
Centre: CRIS
Project website: http://www.csse.monash.edu.au/~karun/Site/research.html
Project outline: Advances in sequencing technologies have revolutionised the genome sequencing landscape, facilitating never before attempted investigations in molecular and cellular biology, and other closely related fields. more...
These new massively-parallel sequencing technologies generate a staggering volume of sequencing “reads” at unprecedented speeds and drastically reduced costs per run.  Large volume of data produced from these new sequencers pose several algorithmic challenges where conventional solutions to various problems simply fail to cope with the scale of data.  This project intends to address several algorithmic and computational challenges associated with Next-Gen sequencing; including the problems of large-scale read-mapping on reference genomes and de novo sequence assembly. less...
Related theme:

Health and wellbeing - Bioinformatics/Protein Structure/Others

Key outcomes:
  1. Konagurthu, A. S.  Allison, L.,  Conway, T., Beresford-Smith, B., & Zobel,J. (2010).  'Design of an efficient out-of-core read alignment algorithm', Tenth international Workshop on Algorithms in Bioinformatics (WABI 2010), Liverpool, UK, 6-8 September, LNCS/LNBI 6293: 189-201.

Adaptiveness of self-organised decision making

Researchers: A/Prof Bernd Meyer
Partners: University of Sydney
Centre: CRIS
Funding: ARC Discovery 2008 -2011
Project Website: http://www.csse.monash.edu.au/~berndm/CDM/
Project outline: Complex systems are an important international research focus in many disciplines, and their engineering applications are plentiful. more...
The new mathematical approach developed by this project will enable different disciplines for the first time to communicate using a common formal framework. This will open the path to a generalised understanding of self-organised systems in dynamic environments. Creating the tools for a unified interdisciplinary approach will allow Australia to gain a stronger position in biomimetic engineering and to take a lead in international research on collective behaviour. less...
Key outcomes:

Details available from the project website above.

Supporting adaptive, interactive documents

Researchers: Prof Kim Marriott, Prof P J Stuckey; Dr B Bos
Partners: University of Melbourne
Centre: CRIS
Funding: ARC Discovery 2009-2011
Project outline: The project will improve comprehensibility of technical material, reduce paper usage, encourage collaborative science, improve the reliability of published science (by allowing post-publication annotation and correction), and improve the accessibility of technical material for readers who are blind or have poor vision. more...

The project also holds considerable potential for supporting Australian companies in the publishing and document processing industries. less...
Key outcomes:  

Dual phase evolution in networks

Researchers: Prof David Green, Prof H A Abass
Partners: ADFA
Centre: CRIS
Funding: ARC Discovery 2009-2011
Project outline: A grand challenge for modern society is the sheer complexity of vast networks arising from organisations and infrastructures. more...
Unexpected, sometimes catastrophic, behaviour often emerges from interactions within such systems. As a result, the Internet, financial markets, power grids and other vital infrastructures are susceptible to costly problems such as cascading failures, inefficiency, and unpredictability. High-tech industries, such as biotechnology and information networking, face problems in coordinating networks of interacting agents. This project will expand the horizon of complex systems by deriving the design principles underpinning stable and resilient network structures and validate these principles on real world networks. less...
Key outcomes:  

Automatic music feature extraction, classification and annotation

Researchers: Prof Guojun Lu, A/Prof Kai Ming Ting, Dr Dengsheng  Zhang
Centre: CRIS
Funding: ARC Discovery 2009-2011
Project outline: Music is a huge industry currently undergoing a major revolution. The industry is shifting from music-making to music retrieval and its incorporation into a range of products from TV, and film, to music streaming into locations and events, as well as MP3 players and all kinds of electronic devices. more...
This research will support immediate retrieval of music that meets the current industry need, based not just on titles, composers and/or performers, but on the actual properties of the music itself. The knowledge and music processing techniques developed will give Australian music industry an advantage over other countries. less...
Key outcomes:  

Multi-sensory fusion and understanding in robotic assistive technology environments

Researchers: Prof Ray Jarvis (Ray.Jarvis@monash.edu),  Prof Ingrid Zukerman (Ingrid.Zukerman@monash.edu)
Centre: CRIS
Funding: ARC Discovery 2008-2010
Project outline: This project integrates spatial and transactional machine intelligence to enable a humanoid robot to interact with people and act on their behalf. more...
The research contributes to the fields of Language Technology and Robotics, addressing the common challenge of interpreting external stimuli. The application domain is assistive technologies for people with limited mobility or cognitive function, thereby extending their independence. less...
Key outcomes: Key outcomes include novel computational models for combining spoken language understanding, vision, touch and sensors to formulate hypotheses about human intent or a situation.

A new paradigm for data modelling

Researchers: A/Prof Kai Ming Ting
Centre: CRIS
Funding: US Air Force Office for Scientific Research and Asia Office of Aerospace Research & Development
Project outline:

The latest research has shown that existing data mining methods have two key limitations: require large data size in order to build a good performing model; and restricted to low dimensional data sets because of high processing time and memory requirements. more...
Our most recent research in anomaly detection tasks have demonstrated that these weaknesses can be overcome by using a fundamentally different approach—mass-based models. This project will deliver the theoretical underpinning of the mass-based approach based on a core modelling technique called mass estimation. The understanding of these theoretical properties will facilitate a wider application of mass estimation to different data mining tasks, especially in classification, regression and clustering. We will show that the mass-based models are trained faster with less data, and they produce better prediction outcomes than existing state-of-the-art methods. Mass estimation represents a paradigm shift in modelling since the introduction of density estimation and support vector machines, with a significant advantage in time and memory requirements that are critical in many real-world applications. less...

Key outcomes:
  • Ting K.M., Zhou, G.T., Liu, F.T., Chuan, T.S. (2010). Mass Estimation and Its Applications. Proceedings of The 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining 2010. pp. 989-998.
  • Liu, T., Ting, K.M. & Zhou, Z-H. (2010). On Detecting Clustered Anomalies using SCiForest. To appear in Proceedings of The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2010).
  • Guang-Tong Zhou, Kai Ming Ting, Fei Tony Liu, Yilong Yin. Relevance Feature Mapping for Content-based Image Retrieval. Proceedings of The 10th International Workshop on Multimedia Data Mining (MDMKDD 2010).
  • Liu, T., Ting, K.M. & Zhou, Z-H. (2008). Isolation Forests. Proceedings of the 2008 IEEE International Conference on Data Mining. 15-19 December 2008. Pisa, Italy, pp. 413-422. [Awarded the Runner-up Best Paper Award in IEEE ICDM 2008]