Discovery Early Career Researcher Award - Grant ID: DE200100964
Funder
Australian Research Council
Funding Amount
$427,068.00
Summary
Early Prediction in Large-scale Time-variant Information Networks. This project aims to develop an early prediction system to predict possible outbreaks of malicious messages in time-variant information networks. The research will primarily leverage deep representations of time-variant subsequence and substructure patterns in large-scale social networks to signal malicious and malevolent messages before it has a chance to propagate. This project will lay the theoretical foundations of this emerg ....Early Prediction in Large-scale Time-variant Information Networks. This project aims to develop an early prediction system to predict possible outbreaks of malicious messages in time-variant information networks. The research will primarily leverage deep representations of time-variant subsequence and substructure patterns in large-scale social networks to signal malicious and malevolent messages before it has a chance to propagate. This project will lay the theoretical foundations of this emerging field to strengthen Australia’s world leadership role in data science. Practically, the novel theories and data analytics technologies developed will help to safeguard Australian business, industry, and society from cyberfraud, online rumour-mongering, and financial loss.Read moreRead less
Enabling Automatic Graph Learning Pipelines with Limited Human Knowledge. This project aims to develop an automatic graph learning system for complex graph data analysis. Machine learning for graph data commonly requires significant human knowledge from both domain professionals as well as algorithm experts, rendering existing systems ineffective and unexplainable. This project expects to design novel graph learning techniques which automatically infer graph relations, learn graph models, adapts ....Enabling Automatic Graph Learning Pipelines with Limited Human Knowledge. This project aims to develop an automatic graph learning system for complex graph data analysis. Machine learning for graph data commonly requires significant human knowledge from both domain professionals as well as algorithm experts, rendering existing systems ineffective and unexplainable. This project expects to design novel graph learning techniques which automatically infer graph relations, learn graph models, adapts existing knowledge to new domains, and provide explanations to the graph learning system. The research results should provide benefit to governments and businesses in many critical applications, such as bioassay activity prediction, credit assessment, and drug discovery and vaccine development in response to the pandemic.Read moreRead less
Deep Pattern Mining for Brain Graph Analysis: A Data Mining Perspective. This project brings together experts in the fields of data mining and cognitive neuroscience. This project aims to develop new data analytics tools, algorithms, and models to combine complex multi-source neuroimage brain data and non-imaging data, to explore the interplays among these different data structures and identify novel functional patterns from complex brain graph structures. The research undertaken in this project ....Deep Pattern Mining for Brain Graph Analysis: A Data Mining Perspective. This project brings together experts in the fields of data mining and cognitive neuroscience. This project aims to develop new data analytics tools, algorithms, and models to combine complex multi-source neuroimage brain data and non-imaging data, to explore the interplays among these different data structures and identify novel functional patterns from complex brain graph structures. The research undertaken in this project expects to provide practical data analysis approaches and establish the theoretical foundations for data mining with multiple sources of brain data.Read moreRead less
Promoting fairness in online attention. This project aims to design mechanisms for fairness of attention to online digital items, by promoting diversity and reducing biases. Attention is one of the most valuable, yet scarce resources in the modern world. Biased attention fuels the propagation of fake information, hurts democratic debate in society and leads to public trust crisis of online media, which could result in unpleasant surprises in individual and group decisions. This project builds ....Promoting fairness in online attention. This project aims to design mechanisms for fairness of attention to online digital items, by promoting diversity and reducing biases. Attention is one of the most valuable, yet scarce resources in the modern world. Biased attention fuels the propagation of fake information, hurts democratic debate in society and leads to public trust crisis of online media, which could result in unpleasant surprises in individual and group decisions. This project builds upon recent breakthroughs in social dynamics, and expects to design new methods for measuring the extent of (un)fairness and (mis)trust, validate novel intervention strategies in a series of online experiments promoting unbiased information consumption and fair decision-making.Read moreRead less
Real-time Event Detection, Prediction, and Visualization for Emergency Response. This project proposes novel end-to-end methods for real-time recognition and prediction of real-world events, leading to timely response to emergencies such as disease outbreaks and natural disasters, as well as prevention of crime, security breaches and the like. It will develop new techniques to quickly detect and predict events by incorporating adaptive learning and probabilistic models, and address fusion and sc ....Real-time Event Detection, Prediction, and Visualization for Emergency Response. This project proposes novel end-to-end methods for real-time recognition and prediction of real-world events, leading to timely response to emergencies such as disease outbreaks and natural disasters, as well as prevention of crime, security breaches and the like. It will develop new techniques to quickly detect and predict events by incorporating adaptive learning and probabilistic models, and address fusion and scalability factors to handle vast collections of heterogeneous data. An event surveillance system prototype will be developed to incorporate the findings of the research with tools to visualise and describe events.Read moreRead less
Design of adaptive learning visual sensor networks for crowd modelling in high-density and occluded scenarios. Partnering University of Melbourne researchers, with video surveillance experts SenSen, engineering consultants ARUP and the Melbourne Cricket Club, the project addresses research enabling a system-integrating, existing surveillance, infrastructure to model crowd behaviour and exit strategies, providing real-time analysis, prediction and response capabilities for venue managers and emer ....Design of adaptive learning visual sensor networks for crowd modelling in high-density and occluded scenarios. Partnering University of Melbourne researchers, with video surveillance experts SenSen, engineering consultants ARUP and the Melbourne Cricket Club, the project addresses research enabling a system-integrating, existing surveillance, infrastructure to model crowd behaviour and exit strategies, providing real-time analysis, prediction and response capabilities for venue managers and emergency services. This new capability enhances utilisation of security resources to prevent injury and fatalities in evacuation scenarios, applicable to existing venues and influencing the development of new facilities around the country. The project delivers researcher training, global clientele for local technology and a platform for local industry growth.Read moreRead less
Effective image search and retrieval through automatic image annotation. This project aims to develop an effective and efficient image retrieval system, so that images are retrieved as easily as textual data. The project researches and develops several key know-hows in image retrieval. It enables Australia to maintain significant advantage in the frontier technology of information processing.
Stream Data Classification in the Age of 5G Networks. This project aims to develop a novel stream data classification model to handle the challenges in the era of 5G networks, such as the scope of the stream data, the complexity of their relationship, the diversity of contained information and the incorrect readings of numerous sensors. The project addresses a significant knowledge gap by exploring and modelling the stronger correlation between data instances in the streams. The outcome is a sys ....Stream Data Classification in the Age of 5G Networks. This project aims to develop a novel stream data classification model to handle the challenges in the era of 5G networks, such as the scope of the stream data, the complexity of their relationship, the diversity of contained information and the incorrect readings of numerous sensors. The project addresses a significant knowledge gap by exploring and modelling the stronger correlation between data instances in the streams. The outcome is a system that is highly efficient, accurate and corrupted-data-tolerant classification solutions for individual stream data as well as multiple stream data. The expected benefits will be far-ranging and adaptable to many domains, such as smart home, medical and healthcare, transportation and manufacturing. Read moreRead less
Scalable Cross-Media Hashing for Searching Heterogeneous Data Sources. This project aims to use novel indexing and search approaches to realise the value of multimedia data. The ever-increasing availability and heterogeneity of multimedia data are calling for new approaches to search information across different media types and data sources. This project aims to enable accurate and real-time cross-media searches from billions of multimedia objects collected from various media sources by tackling ....Scalable Cross-Media Hashing for Searching Heterogeneous Data Sources. This project aims to use novel indexing and search approaches to realise the value of multimedia data. The ever-increasing availability and heterogeneity of multimedia data are calling for new approaches to search information across different media types and data sources. This project aims to enable accurate and real-time cross-media searches from billions of multimedia objects collected from various media sources by tackling the heterogeneity and scalability issues. Expected outcomes include scalable cross-media hashing techniques to capture implicit correlations existing in heterogeneous data and embed high-dimensional features into short binary codes; new binary code indexing and ranking schemes to further improve search speed and quality; and a large-scale cross-media system to evaluate methods and demonstrate the practical value.Read moreRead less
Realising the value of mobile videos with context awareness. Innovative approaches to analysing online video content and context will lead to new ways of interacting with video in the mobile world. This project will aim to develop real-time mobile systems for enabling rich and highly dynamic digital video experiences through context-aware indexing, retrieval and consumption of mobile videos.