Interaction Mining for Cyberbullying Detection on Social Networks. This project plans to build an interactive mining system to detect cyberbullying on social networks that have a large number of participants and a variety of inputs, including conversation texts, time-variant changes and user profiles. The project is designed to change the existing cyberbullying prevention services from reactive keyword filtering to proactive social interaction pattern mining. The intended outcome will enable the ....Interaction Mining for Cyberbullying Detection on Social Networks. This project plans to build an interactive mining system to detect cyberbullying on social networks that have a large number of participants and a variety of inputs, including conversation texts, time-variant changes and user profiles. The project is designed to change the existing cyberbullying prevention services from reactive keyword filtering to proactive social interaction pattern mining. The intended outcome will enable the early detection and warning of cyberbullying and approach open a new way to discover interaction patterns with a large number of participants over evolving and complex social networks.Read moreRead less
Towards interpretable deep learning with limited examples. Existing visual concept detection systems are incapable of detecting ever-evolving concepts in daily life. This project aims to extract patterns that describe the semantics of visual concepts and to develop or adapt knowledge transfer learning technologies for new concepts with limited examples. The expected outcomes will provide major technological breakthroughs for building efficient and interpretable learning systems for visual analys ....Towards interpretable deep learning with limited examples. Existing visual concept detection systems are incapable of detecting ever-evolving concepts in daily life. This project aims to extract patterns that describe the semantics of visual concepts and to develop or adapt knowledge transfer learning technologies for new concepts with limited examples. The expected outcomes will provide major technological breakthroughs for building efficient and interpretable learning systems for visual analysis and will open an entirely new research direction: interpretable deep learning with communication mechanism. This new field and its technologies will help us to recognise misuse of home patient medical devices and unauthorised activity, and enable us to devise effective responses to prevent cyberattacks.Read moreRead less
Data driven decision making for complex problems. This project aims to formulate methods for using constraint solving and data mining in a complementary and holistic manner. Complex health, educational and social issues require complex decisions supported by automated analysis techniques using rich data sources and human knowledge. Constraint solving and data mining make decisions easier, but are mostly deployed independently, limiting the effectiveness of decisions. This project’s methods shoul ....Data driven decision making for complex problems. This project aims to formulate methods for using constraint solving and data mining in a complementary and holistic manner. Complex health, educational and social issues require complex decisions supported by automated analysis techniques using rich data sources and human knowledge. Constraint solving and data mining make decisions easier, but are mostly deployed independently, limiting the effectiveness of decisions. This project’s methods should lead to effective and flexible data driven decision making tools for tackling challenging multi-component problems.Read moreRead less
Multiview Complete Space Learning for Sparse Camera Network Research. Data analytics in video surveillance and social computing is a problem because data are represented by multiple heterogeneous features. This project will develop a multiview complete space learning framework to exploit heterogeneous properties to represent images obtained from sparse camera networks. It will integrate multiple features to identify people and understand behaviour, to build a database of activities occurring in ....Multiview Complete Space Learning for Sparse Camera Network Research. Data analytics in video surveillance and social computing is a problem because data are represented by multiple heterogeneous features. This project will develop a multiview complete space learning framework to exploit heterogeneous properties to represent images obtained from sparse camera networks. It will integrate multiple features to identify people and understand behaviour, to build a database of activities occurring in a wide area of surveillance. It will expand frontier technologies and safeguard Australia by providing warnings for hazardous (for example, overcrowding, trespassing), criminal, and terrorist situations. Results will be applicable internationally and enhance Australia’s role in machine learning and computer vision communities.Read moreRead less
Studying privacy protection methods for multiple independent data releases. Privacy is at risk if two or more published data sets contain overlapping individuals even when each data set is anonymised. This project will investigate if existing anonymisation methods can handle this privacy risk, and will study new solutions. The outcomes will potentially have a great impact on data anonymisation research and applications.
Improving the face of cosmetic medicine - an automatic three-dimensional facial analysis system for facial rejuvenation. 'How will I look?' is the most common question to cosmetic doctors from patients considering facial rejuvenation. This project will answer this question for the first time by providing patients with a three-dimensional model of their post-treatment face as well as informing cosmetic doctors exactly how to achieve the patient's desired face.
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
Machine Learning for Fracture Risk Assessment from Simple Radiography. This project aims to develop a novel, reliable, low-cost system to detect poor bone health and assess fracture risk to help to prevent and manage osteoporosis-related fractures. Currently, osteoporosis-related fractures cost our health system millions of dollars annually and costs are increasing with our ageing population. Early detection of poor bone health will improve the effectiveness of preventive measures and ease this ....Machine Learning for Fracture Risk Assessment from Simple Radiography. This project aims to develop a novel, reliable, low-cost system to detect poor bone health and assess fracture risk to help to prevent and manage osteoporosis-related fractures. Currently, osteoporosis-related fractures cost our health system millions of dollars annually and costs are increasing with our ageing population. Early detection of poor bone health will improve the effectiveness of preventive measures and ease this burden. Current methods include unreliable, crude clinical and visual guides that suggest osteoporosis screening. The project plans to develop a novel system by applying machine learning algorithms to radiology data which is commonly captured for diagnosing other conditions.Read moreRead less
Learning Specific Ontology for Un-Supervised Text Classification. The dramatic rise in massive text data has led to an increasing number of challenges in scalability and noisy information. Supervised classification has become expensive and time consuming as acquiring training sets for a large number of categories becomes more complex and classifiers are sensitive to data. Un-supervised classification has become an attractive alternative given it does not require training sets. However, un-superv ....Learning Specific Ontology for Un-Supervised Text Classification. The dramatic rise in massive text data has led to an increasing number of challenges in scalability and noisy information. Supervised classification has become expensive and time consuming as acquiring training sets for a large number of categories becomes more complex and classifiers are sensitive to data. Un-supervised classification has become an attractive alternative given it does not require training sets. However, un-supervised classification is still complex and there is a gap between understanding of concepts and features. This project aims to exploit domain ontology to find specific ontology which can bridge the gap, leading to a breakthrough for un-supervised classification. It provides foundations for classifying big text data.Read moreRead less
Adversarial Learning of Hybrid Representation. This project aims to design and implement a foundational deep representation learning framework for early detection, classification and defense of emerging malware by capturing their underlying behaviours via structured and unstructured heterogeneous information through hybrid representation learning, behaviour graph mining, and symbolic adversarial learning to discover and defend unknown malware families, thereby significantly boosting the accuracy ....Adversarial Learning of Hybrid Representation. This project aims to design and implement a foundational deep representation learning framework for early detection, classification and defense of emerging malware by capturing their underlying behaviours via structured and unstructured heterogeneous information through hybrid representation learning, behaviour graph mining, and symbolic adversarial learning to discover and defend unknown malware families, thereby significantly boosting the accuracy and robustness of existing classifiers and detectors. The resulting representation learning framework will enhance the national security to protect user privacy, reducing the multi-million-dollar loss caused by fraudulent transactions, and defending against cyber attacks.Read moreRead less