Making Meta-learning Generalised . This project aims to develop novel machine learning techniques, termed generalised meta-learning, to make machines better utilise past experience to solve new tasks with few data. It expects to reduce the undesirable dependence of current machine learning on labelled data and significantly expand its application scope. Expected outcomes of the project consist of new theoretical results on meta-learning and a set of innovative algorithms that can support the bui ....Making Meta-learning Generalised . This project aims to develop novel machine learning techniques, termed generalised meta-learning, to make machines better utilise past experience to solve new tasks with few data. It expects to reduce the undesirable dependence of current machine learning on labelled data and significantly expand its application scope. Expected outcomes of the project consist of new theoretical results on meta-learning and a set of innovative algorithms that can support the building of next generation of computer vision systems to work in open and dynamic environments. This should be able to produce solid benefits to the science, society, and economy of Australian via the application of these advanced intelligent systems.Read moreRead less
Exploring Scientific Information with Advanced New Search Tools. The rapidly growth of scientific literature in many fields makes finding information a challenge. For example, biologists produce over 1 million articles each year. Existing search tools have only limited success satisfying the demands of scientists' queries. This project will deliver intelligent e-research assistants capable of answering scientists' questions directly rather than returning a list of documents. This will allow scie ....Exploring Scientific Information with Advanced New Search Tools. The rapidly growth of scientific literature in many fields makes finding information a challenge. For example, biologists produce over 1 million articles each year. Existing search tools have only limited success satisfying the demands of scientists' queries. This project will deliver intelligent e-research assistants capable of answering scientists' questions directly rather than returning a list of documents. This will allow scientists to more efficiently exploit the literature enabling them to be more innovative and productive. This technology is applicable where ever finding facts in large volumes of text is critical, e.g. analysing surveillance material. Advanced search tools will have considerable academic and industrial impact.Read moreRead less
Parsing the web: Exploiting redundancy to understand language. This project will automatically learn the grammatical structure of language by exploiting redundancy of facts, like 'Mozart was born in 1756', from a trillion words of web text. These facts will be used to understand more complex sentences. This will enable smart information use of text with grammatical information for large-scale information access for the first time. This project will strengthen Australia's world-class expertise, ....Parsing the web: Exploiting redundancy to understand language. This project will automatically learn the grammatical structure of language by exploiting redundancy of facts, like 'Mozart was born in 1756', from a trillion words of web text. These facts will be used to understand more complex sentences. This will enable smart information use of text with grammatical information for large-scale information access for the first time. This project will strengthen Australia's world-class expertise, providing opportunities for future researchers in this area. Our expanded C&C tools and trillion word corpus will be used by academics, companies and governments, in Australia and internationally, aiding applications including financial surveillance and fraud detection.
Read moreRead less
Ask the Net: Intelligent Natural Language Learning. Natural Language Processing (NLP) has progressed rapidly using corpus-based machine learning techniques. However, corpus development costs cause a ?data bottleneck? which prevents systems from reaching human competence. This project overcomes the difficulties of creating huge corpora by employing the innate language ability of untrained contributors. We will show how to automatically select and present examples, containing informative lingui ....Ask the Net: Intelligent Natural Language Learning. Natural Language Processing (NLP) has progressed rapidly using corpus-based machine learning techniques. However, corpus development costs cause a ?data bottleneck? which prevents systems from reaching human competence. This project overcomes the difficulties of creating huge corpora by employing the innate language ability of untrained contributors. We will show how to automatically select and present examples, containing informative linguistic structures, which are most beneficial for training NLP systems. These examples will be analysed by many contributors whose responses will be automatically collated into corpora. Huge corpora are vital to emerging language technologies for managing textual information in the global economy.
Read moreRead less
The DietAdvice Website A New Innovation For Dietitians In Clinical Practice.
Funder
National Health and Medical Research Council
Funding Amount
$140,975.00
Summary
Due to the growing incidence of obesity within Australia, use of computer technology may be a method of targeting these people by increasing access to dietary services. Currently available dietary software in the Australian context only allows analysis of nutrient information. Thus when a dietitian sees a patient they must manually translate food intake to nutrient information, a largely time consuming exercise. DietAdvice is a website that was developed for people to enter in their own food int ....Due to the growing incidence of obesity within Australia, use of computer technology may be a method of targeting these people by increasing access to dietary services. Currently available dietary software in the Australian context only allows analysis of nutrient information. Thus when a dietitian sees a patient they must manually translate food intake to nutrient information, a largely time consuming exercise. DietAdvice is a website that was developed for people to enter in their own food intakes. The food information is sent to a dietitian who develops individualised dietary advice for them. A pilot study of the website has already found it to be feasible in the primary healthcare setting. Tested for 12 months the website was used by 224 patients from GP practices in the Illawarra region of NSW. Approximately 73% of patients were overweight and patients with a high BMI were 1.88 times more likely to use the website in the comfort of their home. Further research about the website however was needed. The research to follow on from the pilot study will aim to refine the DietAdvice website, leading towards its commercialisation for dietitians in clinical practice. The research will be broken into 3 phases. Phase 1 will involve a usability test of the website, assessing the underlying algorithms and testing it with dietitians in private practice. Phase 2 will see volunteers using the website on multiple occasions after being given pre-weighed amounts of food to eat. This will determine how reliable and accurate the information is; and phase 3 will evaluate whether the website is cost effective and if it increases accessibility of health services especially in rural areas. By confirming these attributes there will be a sound basis to commercialise the product.Read moreRead less
Lexical retrieval and reading comprehension: Binding perceptual, lexical and conceptual information in on-line reading. Reading is a complex process that involves integrating sensory information extracted from text with stored memories about word meanings, syntactic structures and general knowledge. Most reading research has focused on the processing of isolated words, but normal reading requires integration processes that are not necessary to recognise single words. This research uses tasks req ....Lexical retrieval and reading comprehension: Binding perceptual, lexical and conceptual information in on-line reading. Reading is a complex process that involves integrating sensory information extracted from text with stored memories about word meanings, syntactic structures and general knowledge. Most reading research has focused on the processing of isolated words, but normal reading requires integration processes that are not necessary to recognise single words. This research uses tasks requiring sentence comprehension and measures of eye movements during reading to investigate how readers retrieve and combine information while reading to comprehend text. It will contribute to developing more comprehensive theories of normal reading that can inform methods of teaching reading and contribute to refinement of text recognition systems.Read moreRead less
Investigations into machine learning applications in link analysis. Link analysis is an emerging tool for the detection of patterns in structured data. The detection of pattern in such data can lead to the detection of fraud occurrence, security breaches in computer systems, and patterns of social interactions with a community. It is also popularly applied to applications such as Web search engine designs and marketing analysis. This project aims to advance the area of link analysis by allowing ....Investigations into machine learning applications in link analysis. Link analysis is an emerging tool for the detection of patterns in structured data. The detection of pattern in such data can lead to the detection of fraud occurrence, security breaches in computer systems, and patterns of social interactions with a community. It is also popularly applied to applications such as Web search engine designs and marketing analysis. This project aims to advance the area of link analysis by allowing the incorporation of contextual information which accounts for relationships among actors properly. Advances in link detection will allow improvements in security and Web services on which a wide field of national bodies rely. This project can help to place Australia at the forefront of this research area.Read moreRead less
Planning, Communication, and Collaboration in Cognitive Systems: A Constructive Approach. Change is a constant and unavoidable characteristic of the current and foreseeable business environment. Currently systems cope poorly with change and as a result they are not sufficiently dependable and adaptable to support business agility and innovation. The aim of this project is to advance the start-of-the art and to lay a new foundation for dependable and adaptable cognitive systems that can plan, com ....Planning, Communication, and Collaboration in Cognitive Systems: A Constructive Approach. Change is a constant and unavoidable characteristic of the current and foreseeable business environment. Currently systems cope poorly with change and as a result they are not sufficiently dependable and adaptable to support business agility and innovation. The aim of this project is to advance the start-of-the art and to lay a new foundation for dependable and adaptable cognitive systems that can plan, communicate and collaborate in complex and dynamic environments.Read moreRead less
Content-based image retrieval using interactive feedback. Storage and retrieval of visual information plays an important role in multimedia systems. We aim to develop a method to retrieve images and visual information from multimedia databases and multimedia systems. The method will go beyond simple similarity search to retrieval data on knowledge by accommodating knowledge acquisition and sequential learning techniques, and will be able to handle complex queries with partial information. The ....Content-based image retrieval using interactive feedback. Storage and retrieval of visual information plays an important role in multimedia systems. We aim to develop a method to retrieve images and visual information from multimedia databases and multimedia systems. The method will go beyond simple similarity search to retrieval data on knowledge by accommodating knowledge acquisition and sequential learning techniques, and will be able to handle complex queries with partial information. The technology developed will find many applications in the multimedia area such as education, medical imaging, desktop publishing, advertising, business presentation, and engineering design.Read moreRead less
A scheme and a formal language for interactive video. Video has been increasingly used in many fields and plays an important role in our daily life. Most traditional video players and video formats only provide linear interaction, such as play, fast forward, fast backward or jump to a certain frame of a video stream indicated by a sequence or time on a random access device. This project aims at enabling a hyperlink type of interactions for video data through the development of video interaction ....A scheme and a formal language for interactive video. Video has been increasingly used in many fields and plays an important role in our daily life. Most traditional video players and video formats only provide linear interaction, such as play, fast forward, fast backward or jump to a certain frame of a video stream indicated by a sequence or time on a random access device. This project aims at enabling a hyperlink type of interactions for video data through the development of video interaction mark-up language (VIML). It can find many applications in education, marketing, web presentation and digital TV.Read moreRead less