Computing with nearly-consistent data. This project will help programmers correctly use data that originates at various times and places, and spreads unevenly through a system. Computation will combine data that comes from different situations, and is not exactly consistent. Capability to develop high quality software on platforms with this feature will enhance the value of the Australian IT industry. As well, the industries which use the software benefit from correctly working with their data. ....Computing with nearly-consistent data. This project will help programmers correctly use data that originates at various times and places, and spreads unevenly through a system. Computation will combine data that comes from different situations, and is not exactly consistent. Capability to develop high quality software on platforms with this feature will enhance the value of the Australian IT industry. As well, the industries which use the software benefit from correctly working with their data. Sensor networks have data like this, and they play a vital role in environmental monitoring. Cloud computing platforms also have this type of data, and these allow smaller enterprises to grow smoothly, without needing large up-front investments in computing infrastructure.Read moreRead less
Adaptive Key-value Store for Future Extreme Heterogeneous Systems. Safe, lasting storage of data, and efficient access to it, is vital for all aspects of computing, ranging from e-commerce applications, and data-management in governments. For the storage of data, persistent key-value stores are central in modern computing platforms. However, contemporary key-value stores have not been designed for emerging extreme heterogeneous computational systems with future hardware accelerators and storage ....Adaptive Key-value Store for Future Extreme Heterogeneous Systems. Safe, lasting storage of data, and efficient access to it, is vital for all aspects of computing, ranging from e-commerce applications, and data-management in governments. For the storage of data, persistent key-value stores are central in modern computing platforms. However, contemporary key-value stores have not been designed for emerging extreme heterogeneous computational systems with future hardware accelerators and storage capabilities, including graphics processor and flash-based memory. This project will devise an adaptive key-value store framework for heterogeneous systems. Our new framework will adaptively harvest the performance potential of future hardware such that applications can cope with fast-growing data sets.Read moreRead less
What Can You Trust in the Large and Noisy Web? This project will develop innovative techniques to efficiently and effectively distill truthful information from the inherently unreliable and large-scale Web environment, where misinformation has been widely regarded as a grand challenge for the next decade. The results of this project will not only maintain Australia’s leadership in this frontier research area, but also support many important applications that safeguard Australian people and econo ....What Can You Trust in the Large and Noisy Web? This project will develop innovative techniques to efficiently and effectively distill truthful information from the inherently unreliable and large-scale Web environment, where misinformation has been widely regarded as a grand challenge for the next decade. The results of this project will not only maintain Australia’s leadership in this frontier research area, but also support many important applications that safeguard Australian people and economy such as emergency and disaster management and online healthcare. This project also serves as an excellent vehicle for the education and training of Australia’s next generation of scholars and engineers.Read moreRead less
Improving provision of a document store as a service in a public cloud. Improving provision of a document store as a service in a public cloud. This project aims to develop a model of document-oriented database correctness and performance that can be applied to cloud-hosted clusters. Many modern web applications rely on document-oriented databases hosted on clusters of virtualised servers from commercial cloud providers. Developers make difficult deployment decisions, such as how to combine data ....Improving provision of a document store as a service in a public cloud. Improving provision of a document store as a service in a public cloud. This project aims to develop a model of document-oriented database correctness and performance that can be applied to cloud-hosted clusters. Many modern web applications rely on document-oriented databases hosted on clusters of virtualised servers from commercial cloud providers. Developers make difficult deployment decisions, such as how to combine data sharding and replication to meet service requirements without any guidance on the degree of precision. This research will allow developers and database administrators to predict how systems will behave in conditions difficult to simulate directly. Ultimately, this is expected to improve the quality and efficiency of services built using document databases.Read moreRead less
Efficient Techniques for Mining Exceptional Patterns. This research will develop totally new techniques for exceptional pattern discovery that are useful for deeper understanding data mining and capturing the hidden interactions (class-bridge rules and out-expectation patterns) within data. This will enable Australian data marketers to access valuable implicit information that is contained in their data, but not currently accessible. The outcomes will keep Australia in the international leading ....Efficient Techniques for Mining Exceptional Patterns. This research will develop totally new techniques for exceptional pattern discovery that are useful for deeper understanding data mining and capturing the hidden interactions (class-bridge rules and out-expectation patterns) within data. This will enable Australian data marketers to access valuable implicit information that is contained in their data, but not currently accessible. The outcomes will keep Australia in the international leading edge and preserve its competitive status in preemptively defining the information market of tomorrow. To 'Frontier Technologies for Building and Transforming Australian Industries', discovering new exceptional patterns within data will lead to increased efficiency in Australian Industries.Read moreRead less
Developing an effective defence to cyber-reputation manipulation attacks. This project will develop new technologies for businesses to accurately identify fake internet reviews. Fake reviews, paid for and/or written with malicious intent, can cause irreparable damage to businesses resulting in revenue loss, consumer dissatisfaction or even closure of businesses. However they are difficult to identify, as they continuously evolve to avoid detection and the volume of Internet reviews makes analysi ....Developing an effective defence to cyber-reputation manipulation attacks. This project will develop new technologies for businesses to accurately identify fake internet reviews. Fake reviews, paid for and/or written with malicious intent, can cause irreparable damage to businesses resulting in revenue loss, consumer dissatisfaction or even closure of businesses. However they are difficult to identify, as they continuously evolve to avoid detection and the volume of Internet reviews makes analysis a monumental task. This project will provide advanced tools to detect fake website reviews and a cybersecurity system prototype ready to be used by industry, making Australia a leader in this field and resulting in a safer internet environment for all.Read moreRead less
Efficient Strategies for Mining Negative Association Rules. Negative association rules (NAR) catch mutually-exclusive correlations
among items. They play important roles just as traditional association
rules (TAR) do. For example, in stock market surveillance based on alert logs, NARs detect which alerts are false. There are essential differences between mining TARs and NARs because NARs are hidden in infrequent itemsets. This research will develop efficient strategies for mining NARs in datab ....Efficient Strategies for Mining Negative Association Rules. Negative association rules (NAR) catch mutually-exclusive correlations
among items. They play important roles just as traditional association
rules (TAR) do. For example, in stock market surveillance based on alert logs, NARs detect which alerts are false. There are essential differences between mining TARs and NARs because NARs are hidden in infrequent itemsets. This research will develop efficient strategies for mining NARs in databases. These strategies are expected to be about ten times faster than existing ones. This project will deliver database-independent and high-performance mining algorithms for decision-making. The results can benefit Australian marketing and financial companies as well as health and security departments for smart information use.Read moreRead less
Multiple Data Source Discovery: Group Interaction Approach. This project will develop new technology and theory to identify and evaluate incomplete data. It will deliver a high-performance group-interaction based global pattern discovery system that enables decision-makers (like doctors) to access valuable implicit information that is contained in their data but not currently accessible. Mining group interactions will greatly extend the scope of pattern discovery and new product evaluation. The ....Multiple Data Source Discovery: Group Interaction Approach. This project will develop new technology and theory to identify and evaluate incomplete data. It will deliver a high-performance group-interaction based global pattern discovery system that enables decision-makers (like doctors) to access valuable implicit information that is contained in their data but not currently accessible. Mining group interactions will greatly extend the scope of pattern discovery and new product evaluation. The outcomes of the project will lead to better diagnostic decisions and will lead to increased efficiency in Australian Industries.Read moreRead less
Effective Techniques and Methodologies for Multi-Database Mining. This project develops a high-performance multi-database mining system. This project is significant because (1) it is imperative due to a great deal of multi-databases widely used in organizations; (2) it is difficult due to essential differences between mono- and multi-databases; (3) existing multi-database mining techniques are inadequate; and (4) the new mining strategies in this project can make a vast improvement of the abilit ....Effective Techniques and Methodologies for Multi-Database Mining. This project develops a high-performance multi-database mining system. This project is significant because (1) it is imperative due to a great deal of multi-databases widely used in organizations; (2) it is difficult due to essential differences between mono- and multi-databases; (3) existing multi-database mining techniques are inadequate; and (4) the new mining strategies in this project can make a vast improvement of the ability and performance of multi-database mining systems. The expected outcomes are: an application-independent database classification, a local instance analysis and a prototype system. These proposed techniques are innovative, effective and efficient in identifying novel patterns from multi-databases.Read moreRead less
Ontology-Based Group Pattern Discovery Systems for Mining Multiple Data Sources. This project will aim at the frontier technologies development for practical techniques in the context of real multiple-data-source mining systems, including stock data and e-business data analysis. It will bring Australian individuals and organizations (i) high quality information from multiple data sources and (ii) automatically pattern discovery systems for tackling the multiple data source problem. This will lea ....Ontology-Based Group Pattern Discovery Systems for Mining Multiple Data Sources. This project will aim at the frontier technologies development for practical techniques in the context of real multiple-data-source mining systems, including stock data and e-business data analysis. It will bring Australian individuals and organizations (i) high quality information from multiple data sources and (ii) automatically pattern discovery systems for tackling the multiple data source problem. This will lead to greatly enhance the international competition of Australian companies and significantly reduce investing risks. Read moreRead less