Data Mining by Clustering in Very Large Relational Databases. Many commercial and governmental entities possess very large relational data that cannot be feasibly analyzed by today's computers, e.g., gene expression data, product usage databases and telecommunication call records. The clustering tools developed in this project will have a significant benefit on many business processes that involve clustering this type of data, such as fraud detection and market segmentation.
Emergence of robust, stable structures via computation within natural networks. An ever-increasing challenge for modern society is the sheer complexity of vast infrastructures. Unexpected, and sometimes catastrophic, behaviour often emerges from interactions between elements of large systems. As a result, highly complex systems such as the Internet, international finance markets, and power grids are highly susceptible to costly problems such as cascading failures, inefficiency, and critical sens ....Emergence of robust, stable structures via computation within natural networks. An ever-increasing challenge for modern society is the sheer complexity of vast infrastructures. Unexpected, and sometimes catastrophic, behaviour often emerges from interactions between elements of large systems. As a result, highly complex systems such as the Internet, international finance markets, and power grids are highly susceptible to costly problems such as cascading failures, inefficiency, and critical sensitivity. High-tech industries, such as biotechnology and information networking, also face problems in coordinating swarms of interacting agents. This project will contribute to solving such problems by identifying and adapting solutions from nature.Read moreRead less
Dual phase evolution in networks. A grand challenge for modern society is the sheer complexity of vast networks arising from organizations and infrastructures. 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 info ....Dual phase evolution in networks. A grand challenge for modern society is the sheer complexity of vast networks arising from organizations and infrastructures. 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.Read moreRead less
Managing private location data in a mobile and networked world: getting the balance right. Location based data are transforming the mobile service industry and this project will develop novel approaches to safeguard the location privacy of mobile individuals. This will facilitate the development of privacy-aware services which can be used for real time traffic monitoring, care for the elderly and smartphone enabled location services.
Long-term Cloud Service Composition. This project proposes an economic model-based framework for the selection and composition of cloud services, thus creating an efficient market for cloud consumers and providers. The project will use economic models that incorporate a range of quality of service (QoS) parameters as a key driver for optimising the selection of cloud services and the acceptance of consumer requests. The main outcomes of this project aim to increase efficiencies in the cloud mark ....Long-term Cloud Service Composition. This project proposes an economic model-based framework for the selection and composition of cloud services, thus creating an efficient market for cloud consumers and providers. The project will use economic models that incorporate a range of quality of service (QoS) parameters as a key driver for optimising the selection of cloud services and the acceptance of consumer requests. The main outcomes of this project aim to increase efficiencies in the cloud market, benefiting consumers and providers.Read moreRead less
Developing Adversary-Aware Classifiers for Android Malware Detection. Smartphones have become increasingly ubiquitous in people’s everyday life. However, it was reported that one in every five Android applications were actually malware, considering that Android has taken 88% market share of mobile phones. As an effective technique, machine learning has been widely adopted to detect Android malware. However, recent work suggests that deliberately-crafted malware makes machine learning ineffective ....Developing Adversary-Aware Classifiers for Android Malware Detection. Smartphones have become increasingly ubiquitous in people’s everyday life. However, it was reported that one in every five Android applications were actually malware, considering that Android has taken 88% market share of mobile phones. As an effective technique, machine learning has been widely adopted to detect Android malware. However, recent work suggests that deliberately-crafted malware makes machine learning ineffective. In this project, we propose to develop a series of new techniques, such as 1) Android contextual analysis, 2) wrapper-based hill climbing algorithm, and 3) ensemble learning, to solve this problem. The outcomes will help Australia gain cutting edge technologies in adversarial machine learning and mobile security.Read moreRead less
Tracing real Internet attackers through information correlation. If this research accomplishes successfully, it will be a big step forward on tracing Internet attackers in terms of traceback scope, accuracy, usability and deployment. This will empower authorities to control and punish Internet crime and terrorism. It will also greatly reduce the damage caused by Internet crime and terrorism. The prototype of the distributed information correlation tracing system can possibly be patented or even ....Tracing real Internet attackers through information correlation. If this research accomplishes successfully, it will be a big step forward on tracing Internet attackers in terms of traceback scope, accuracy, usability and deployment. This will empower authorities to control and punish Internet crime and terrorism. It will also greatly reduce the damage caused by Internet crime and terrorism. The prototype of the distributed information correlation tracing system can possibly be patented or even be commercialised. The capability of a nation to trace the real source of any attacks on its information infrastructure is central to the control of such attacks and hence to a nation's long-term survival and prosperity.Read moreRead less
Growing old and staying connected: touch screen technology for ameliorating older people’s experience of social isolation. Social isolation affects many older people. This project investigates novel technologies to prevent and to ameliorate social isolation experienced by older adults. This project will implement and trial a software application over an 18 month period, using a 3G connected touch-screen tablet, and evaluate its impact on alleviating social isolation.
Classifying Internet traffic for security applications. As the internet traffic data exponentially increases every year, traffic classification has become a fundamental approach to the security of the Internet. This project aims to develop a set of novel techniques for internet traffic classification, which is fundamentally important to defend against the serious cyber-attacks and effectively minimise the damages. This project is significant as it can help to improve cyber security, which is ess ....Classifying Internet traffic for security applications. As the internet traffic data exponentially increases every year, traffic classification has become a fundamental approach to the security of the Internet. This project aims to develop a set of novel techniques for internet traffic classification, which is fundamentally important to defend against the serious cyber-attacks and effectively minimise the damages. This project is significant as it can help to improve cyber security, which is essential for the work and daily lives of the Australian people. Furthermore, the proposed models and techniques will be important for enhancing the protection of Australian critical infrastructures against malicious cyber-attacks.Read moreRead less
Data-driven Traffic Analytics for Incident Analysis and Management. Traffic incidents are among the primary concerns of all transport authorities around the world due to their significant impact in terms of traffic congestion and delay, air and noise pollution, and management cost. This project aims to address incident analysis and management in complex and multi-modal traffic networks by combining multidisciplinary research efforts from transportation engineering and data science. The intended ....Data-driven Traffic Analytics for Incident Analysis and Management. Traffic incidents are among the primary concerns of all transport authorities around the world due to their significant impact in terms of traffic congestion and delay, air and noise pollution, and management cost. This project aims to address incident analysis and management in complex and multi-modal traffic networks by combining multidisciplinary research efforts from transportation engineering and data science. The intended outcomes will be an innovative incident analysis and management framework synergising traffic data analytics and traffic simulation modelling as well as its key enabling techniques and prototype systems. This will significantly help mitigate incident impacts on daily commuters.Read moreRead less