Cohort discovery and activity mining for policy impact prediction. Cohort discovery and activity mining for policy impact prediction. This project aims to develop an intelligent systematic framework to predict policy impacts on Australian patients, by discovering inherent patient cohorts and assessing the impact of the policies on these cohorts. The proposed methods lay the theoretical foundations for building intelligent automated tools for policy assessment. Expected outcomes are data-driven p ....Cohort discovery and activity mining for policy impact prediction. Cohort discovery and activity mining for policy impact prediction. This project aims to develop an intelligent systematic framework to predict policy impacts on Australian patients, by discovering inherent patient cohorts and assessing the impact of the policies on these cohorts. The proposed methods lay the theoretical foundations for building intelligent automated tools for policy assessment. Expected outcomes are data-driven patient group discovery, which could more precisely identify the patient cohorts most likely to benefit from a specific policy; and a model to predict the efficacy of policy options, which could increase the sustainability of the national health system by enabling smarter, more efficient policy decision-making.Read moreRead less
Intruder alert! detecting and classifying events in noisy time series. This project aims to address the mathematical challenges in automated early detection and classification of intrusion events in noisy time series generated from perimeter security systems. The project expects to develop robust methods to detect intrusion events under different operating environments while ignoring nuisance events. The project will boost the global competitiveness of the Australian security industry, and enabl ....Intruder alert! detecting and classifying events in noisy time series. This project aims to address the mathematical challenges in automated early detection and classification of intrusion events in noisy time series generated from perimeter security systems. The project expects to develop robust methods to detect intrusion events under different operating environments while ignoring nuisance events. The project will boost the global competitiveness of the Australian security industry, and enable improved event detection and classification in noisy time series to the benefit of many critical application areas beyond national security.Read moreRead less
Automatic speech-based assessment of mental state via mobile device. This project aims to create the first mobile, device-based automatic assessment of mental state from acoustic speech. Focusing on novel approaches for eliciting speech, for regression-based scoring of mental state and for longitudinal modelling of speech, the project takes speech processing out of the laboratory and into realistic environments. The project is significant because elicitation approach and longitudinal modelling h ....Automatic speech-based assessment of mental state via mobile device. This project aims to create the first mobile, device-based automatic assessment of mental state from acoustic speech. Focusing on novel approaches for eliciting speech, for regression-based scoring of mental state and for longitudinal modelling of speech, the project takes speech processing out of the laboratory and into realistic environments. The project is significant because elicitation approach and longitudinal modelling have been acknowledged by the research community as challenges that are valuable to investigate, and because conventional regression methods are sub-optimal on ordinal mental state scales. This is significant commercially because mobile devices allow individually tailored, frequent and low-cost mental state assessment. Expected outcomes will include commercial-ready technology, trialled on Australians, accessible to everyone with a mobile device and concentration of Australian research and development capability in a rapidly growing application area.Read moreRead less
Privacy-preserving cloud data mining-as-a-service. This project aims to explore practical privacy-preserving solutions for cloud data mining-as-a-service based on the Intel Software Guard Extensions (SGX) technology. The research addresses privacy concerns of users when outsourcing data mining needs to the cloud. These concerns have increased as more businesses evaluate data mining-as-an outsourced service due to lack of expertise or computation resources. The expected outcomes from the research ....Privacy-preserving cloud data mining-as-a-service. This project aims to explore practical privacy-preserving solutions for cloud data mining-as-a-service based on the Intel Software Guard Extensions (SGX) technology. The research addresses privacy concerns of users when outsourcing data mining needs to the cloud. These concerns have increased as more businesses evaluate data mining-as-an outsourced service due to lack of expertise or computation resources. The expected outcomes from the research will include new data privacy models, new privacy-preserving data mining algorithms, and a prototype of cloud data mining software. These will help businesses cut costs for data mining and privacy protection, and provide significant benefits toward helping Australia achieve its national cyber security strategy and potentially provide economic impact from commercialisation of new software technology for the industry partner.Read moreRead less
Mining large negative correlations for high-dimensional contrasting analysis. Negative correlations are widely embedded in real life applications, but in-depth research has rarely been conducted due to its high level of complexity. This project aims at efficient algorithms and frontier theory for finding large negative correlations, to enable smart information use in bioinformatics to promote Australia's leading role in data mining research.
Discovery Early Career Researcher Award - Grant ID: DE120101161
Funder
Australian Research Council
Funding Amount
$375,000.00
Summary
Compressive sensing based probabilistic graphical models (PGM). The aim of the project is to develop fast, large scale probabilistic graphical models (PGM) learning and inference methods. The resulting system will be able to process large scale PGMs on a standard PC, and will be easily extendable to computer clustering for larger scale PGMs requiring higher precision.
Privacy Preserving Data Sharing in Electronic Health Environment. This project aims to improve access to electronic health data (EHD) while still ensuring patient privacy. EHD can provide important information for medical research and health-care resource allocations. However, data sharing in electronic health environments is challenging because of the privacy concerns of customers. Large-scale unauthorised access from internal staff has been reported in Medicare. This project aims to develop ne ....Privacy Preserving Data Sharing in Electronic Health Environment. This project aims to improve access to electronic health data (EHD) while still ensuring patient privacy. EHD can provide important information for medical research and health-care resource allocations. However, data sharing in electronic health environments is challenging because of the privacy concerns of customers. Large-scale unauthorised access from internal staff has been reported in Medicare. This project aims to develop new privacy-preserving algorithms on EHD database federations, which can provide efficient data access yet block inside attacks. It will significantly improve the data available for medical research, while reducing the cost of EHD system management and providing visualised decision supports to medical staff and the government health resource planners.Read moreRead less
Efficient causal discovery from observational data. Discovering cause-effect relationships is the ultimate goal for many applications. Randomised control trial is the gold standard for discovering causal relationships. However, conducting such trials is impossible in many cases due to cost and/or ethical concerns. In contrast, a large amount of data has been accumulated in all areas. It is desirable to infer causal relationships from data directly and automatically. This project aims to develop ....Efficient causal discovery from observational data. Discovering cause-effect relationships is the ultimate goal for many applications. Randomised control trial is the gold standard for discovering causal relationships. However, conducting such trials is impossible in many cases due to cost and/or ethical concerns. In contrast, a large amount of data has been accumulated in all areas. It is desirable to infer causal relationships from data directly and automatically. This project aims to develop fast and scalable data mining methods for identifying causal relationships from large and/or high dimensional data sets. The developed methods will mainly be evaluated in real world biological applications. The research outcomes will be useful in many areas for causal reasoning and decision making.Read moreRead less
Developing novel data mining methods to reveal complex group relationships from heterogeneous data. This project aims to develop novel and effective data mining methods that will enable us to unravel the relationships between multiple, rather than individual, components of complex systems (such as genes, gene regulators and cancer), which is crucial to understanding how such systems work. Potential applications for such methods are extensive.
Online Learning for Large Scale Structured Data in Complex Situations. Online Learning (OL) is the process of predicting answers for a sequence of questions. OL has enjoyed much attention in recent years due to its natural ability of processing large scale non-structured data and adapting to a changing environment. However, OL has three weaknesses: it does not scale for structured data; it often assumes that all of the data are equally important; it often considers that all of the data are compl ....Online Learning for Large Scale Structured Data in Complex Situations. Online Learning (OL) is the process of predicting answers for a sequence of questions. OL has enjoyed much attention in recent years due to its natural ability of processing large scale non-structured data and adapting to a changing environment. However, OL has three weaknesses: it does not scale for structured data; it often assumes that all of the data are equally important; it often considers that all of the data are complete and noise-free. These weaknesses limit its utility, because real data such as those that must be analysed in processing social networks, fraud detection do not satisfy the restrictions. The aim of this project is to develop theoretical and practical advances in OL that overcome the existing weaknesses.Read moreRead less