Discovery Early Career Researcher Award - Grant ID: DE210101458
Funder
Australian Research Council
Funding Amount
$387,141.00
Summary
Scalable and Deep Anomaly Detection from Big Data with Similarity Hashing. Anomaly detection, aiming to identify anomalous but insightful patterns in data mining, is an important big data analytics technique. The nature of big data requires a detection method that can handle fast-evolving data of diverse types. However, existing methods suffer from either high computational cost or low detection performance. This project aims to develop a detection framework to advance detection performance and ....Scalable and Deep Anomaly Detection from Big Data with Similarity Hashing. Anomaly detection, aiming to identify anomalous but insightful patterns in data mining, is an important big data analytics technique. The nature of big data requires a detection method that can handle fast-evolving data of diverse types. However, existing methods suffer from either high computational cost or low detection performance. This project aims to develop a detection framework to advance detection performance and efficiency, based on a novel deep learning model called deep isolation forest which is different from the traditional artificial neural network based models. The outcome will bring huge benefits to various applications such as real-time predictive maintenance in smart manufacturing, and intrusion detection in cybersecurity.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE190101118
Funder
Australian Research Council
Funding Amount
$339,000.00
Summary
High performance density-based clustering in parallel environments. This project aims to conduct a comprehensive study on density-based clustering to improve data management in parallel computing environments. Clustering, a fundamental task in data management, is to group a set of objects such that objects in the same group (called a cluster) are more similar to each other than those in other groups in order to simplify retrieval of similar information. Clustering is widely used in many fields i ....High performance density-based clustering in parallel environments. This project aims to conduct a comprehensive study on density-based clustering to improve data management in parallel computing environments. Clustering, a fundamental task in data management, is to group a set of objects such that objects in the same group (called a cluster) are more similar to each other than those in other groups in order to simplify retrieval of similar information. Clustering is widely used in many fields including machine learning, pattern recognition, information retrieval, bioinformatics and image analysis. It is expected that the developed clustering techniques will provide significant performance improvements in industry sectors where decisions are made based on clustering data analytics, such as the sectors of finance, renewable energy and artificial intelligence.Read moreRead less
Advancing Analytical Query Processing with Urban Trajectory Data. This project aims to provide accurate, rapid, and comprehensive information to analyze transport and related infrastructure use in real time. This project expects to develop innovative solutions by exploiting massive urban trajectory data derived from public transport usage, route mapping, GPS tracking and road-side sensors. Expected outcomes include a new algorithmic framework to support complex trajectory-driven analytical tasks ....Advancing Analytical Query Processing with Urban Trajectory Data. This project aims to provide accurate, rapid, and comprehensive information to analyze transport and related infrastructure use in real time. This project expects to develop innovative solutions by exploiting massive urban trajectory data derived from public transport usage, route mapping, GPS tracking and road-side sensors. Expected outcomes include a new algorithmic framework to support complex trajectory-driven analytical tasks in public transport network planning, traffic congestion prevention, and facility deployment. This should significantly benefit both government and industry in data-driven decision makings and evaluations on the impact of decisions made, and ultimately materialize Australian government’s Smart Cities Plan.Read moreRead less
Decentralised Collaborative Predictive Analytics on Personal Smart Devices. This project tackles the challenging problem of personalised predictive analytics with resource-constrained personal devices and massive-scale data. The knowledge to be generated concerns privacy, fairness, and resource efficiency in the era of Internet of Things. The expected outcomes include a collaborative learning paradigm for building personalised models on personal smart devices in open and fully decentralised sett ....Decentralised Collaborative Predictive Analytics on Personal Smart Devices. This project tackles the challenging problem of personalised predictive analytics with resource-constrained personal devices and massive-scale data. The knowledge to be generated concerns privacy, fairness, and resource efficiency in the era of Internet of Things. The expected outcomes include a collaborative learning paradigm for building personalised models on personal smart devices in open and fully decentralised settings. Privacy and model fairness are core tenets of the paradigm. Personalised predictive analytics is frontier research that will position Australia at the forefront of AI and give business the tools needed to deploy innovative business systems for market exploitation with a secure, equitable and competitive advantage.Read moreRead less
Making Spatiotemporal Data More Useful: An Entity Linking Approach. This project aims to establish a methodology for spatiotemporal entity linking by utilising object movement traces to support database integration and data quality management for the next-generation of data where spatiotemporal attributes are ubiquitous. It expects to develop a novel entity linking paradigm for automatic, efficient and reliable spatiotemporal data integration together with a new data privacy study in this contex ....Making Spatiotemporal Data More Useful: An Entity Linking Approach. This project aims to establish a methodology for spatiotemporal entity linking by utilising object movement traces to support database integration and data quality management for the next-generation of data where spatiotemporal attributes are ubiquitous. It expects to develop a novel entity linking paradigm for automatic, efficient and reliable spatiotemporal data integration together with a new data privacy study in this context. Expected outcome include new database technologies for data signature generation and similarity-based search, and improved location data privacy protection methods. This project should provide significant benefits to all areas where high quality spatiotemporal data fusion is essential to meaningful data analysis.Read moreRead less
A ubiquitous system for indoor location-based services. This project aims to design a system capable of providing location-based services in WiFi-enabled indoor buildings. The project expects to address two major challenges hindering such a system, capability to identify a user's location in any indoor venue with minimum manual overhead, and effective data management techniques for indoor data. Expected outcomes of this project include new techniques for managing and utilising indoor location da ....A ubiquitous system for indoor location-based services. This project aims to design a system capable of providing location-based services in WiFi-enabled indoor buildings. The project expects to address two major challenges hindering such a system, capability to identify a user's location in any indoor venue with minimum manual overhead, and effective data management techniques for indoor data. Expected outcomes of this project include new techniques for managing and utilising indoor location data and enhanced international collaborations. The project will support and enhance a wide range of indoor applications such as emergency services, assisted healthcare systems, indoor asset tracking, event planning, indoor venue management, and consumer experience.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE210100160
Funder
Australian Research Council
Funding Amount
$423,000.00
Summary
Information Extraction from Large-scale Low-quality Data. Information extraction which identifies entities and relations from data is a key technology that lays the foundation for understanding the semantics of data. This project aims to investigate the problem of information extraction by innovatively exploring the informality and temporal evolution of data. It expects to develop novel techniques for reliable, efficient, and scalable information discovery from large-scale low-quality data. Expe ....Information Extraction from Large-scale Low-quality Data. Information extraction which identifies entities and relations from data is a key technology that lays the foundation for understanding the semantics of data. This project aims to investigate the problem of information extraction by innovatively exploring the informality and temporal evolution of data. It expects to develop novel techniques for reliable, efficient, and scalable information discovery from large-scale low-quality data. Expected outcomes include a set of collective, contextualised, and temporal-aware algorithms for information extraction and integration, built on top of effective indexing and in-parallel processing. This project is anticipated to benefit a considerable number of data-driven intelligence-based applications.Read moreRead less
Efficient and Scalable Similarity Query Processing on Big Streaming Graphs . This project aims to develop novel approaches for efficient and scalable similarity queries on big streaming graphs which are large-scale graphs where graph nodes and edges may arrive or expire at high speed. Three key challenges are expected to be addressed including high speed, large variety, and big volume of streaming graphs. Expected outcomes include new theories, novel indexing and query processing techniques, an ....Efficient and Scalable Similarity Query Processing on Big Streaming Graphs . This project aims to develop novel approaches for efficient and scalable similarity queries on big streaming graphs which are large-scale graphs where graph nodes and edges may arrive or expire at high speed. Three key challenges are expected to be addressed including high speed, large variety, and big volume of streaming graphs. Expected outcomes include new theories, novel indexing and query processing techniques, and advanced distributed algorithms as well as a system prototype for evaluation and to demonstrate the practical value. Success in this project should see significant benefits for many important applications, such as e-commerce, cybersecurity, health, social networks, and bio-informatics.
Read moreRead less
Taming Large-Volume Dynamic Graphs in the Cloud. This project aims to develop efficient and scalable algorithms to process large-volume dynamic graphs in the cloud. The project expects to address key challenges and lay theoretical foundations in large-volume dynamic graph processing, which plays an important role in developing general-purpose, real-time structural search engines. Expected outcomes of this project include theoretical foundations and scalable algorithms to process big graphs that ....Taming Large-Volume Dynamic Graphs in the Cloud. This project aims to develop efficient and scalable algorithms to process large-volume dynamic graphs in the cloud. The project expects to address key challenges and lay theoretical foundations in large-volume dynamic graph processing, which plays an important role in developing general-purpose, real-time structural search engines. Expected outcomes of this project include theoretical foundations and scalable algorithms to process big graphs that evolve rapidly over time. These enable users to monitor and analyse structural information in large dynamic networks in real time. The project expects to open up a new research direction for graph processing to enrich frontier technologies and benefit many key applications in Australia.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE200101465
Funder
Australian Research Council
Funding Amount
$419,498.00
Summary
Minimising Human Efforts to Fight Fake News and Restore the Public Trust. Our modern society is struggling with an unprecedented amount of online fake news, which is recently driven by misused artificial intelligence (AI) technologies. This project aims to build the first real-time system integrating algorithmic models and human validators to counter such falsehoods, especially those AI-fabricated false stories. This project expects to deliver a series of cost-effective and streaming methods emp ....Minimising Human Efforts to Fight Fake News and Restore the Public Trust. Our modern society is struggling with an unprecedented amount of online fake news, which is recently driven by misused artificial intelligence (AI) technologies. This project aims to build the first real-time system integrating algorithmic models and human validators to counter such falsehoods, especially those AI-fabricated false stories. This project expects to deliver a series of cost-effective and streaming methods empowering a Web-based observatory dashboard of fake news propagation. This achieves significant benefits for media organisations, governments, the public, and academia via timely alerts, data-journalism reports, and novel data visualisations of social media landscape to distinguish between legitimate and deceptive contents.Read moreRead less