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.
Efficient and effective ad-hoc search using structured and unstructured geospatial information. Web search is a key enabling technology in the information age. However two technologies, ubiquitous mobile devices and massive structured data repositories such as those used to maintain social networking sites, are changing user expectations about how and what should be searched. A key challenge in the research community is how to integrate structured and unstructured information to improve the qual ....Efficient and effective ad-hoc search using structured and unstructured geospatial information. Web search is a key enabling technology in the information age. However two technologies, ubiquitous mobile devices and massive structured data repositories such as those used to maintain social networking sites, are changing user expectations about how and what should be searched. A key challenge in the research community is how to integrate structured and unstructured information to improve the quality of search. This project proposes new approaches to ranked retrieval for location-aware search. In particular, it presents a plan to combine state-of-the-art research from two domains: spatial keyword search in databases, and ad-hoc search in Information Retrieval to improve the quality of search results.Read moreRead less
User-Adaptive Search and Evaluation for Complex Information-Seeking Tasks. This project plans to develop a new evaluation framework to understand and characterise web users and their situation within complex, multi-faceted search tasks, exemplified through job-search. While evaluation of web search engine effectiveness is relatively well understood, measuring information retrieval performance in the context of complex tasks with heterogeneous users is a largely neglected problem. This project pl ....User-Adaptive Search and Evaluation for Complex Information-Seeking Tasks. This project plans to develop a new evaluation framework to understand and characterise web users and their situation within complex, multi-faceted search tasks, exemplified through job-search. While evaluation of web search engine effectiveness is relatively well understood, measuring information retrieval performance in the context of complex tasks with heterogeneous users is a largely neglected problem. This project plans to mine user-specific characteristics and situations from complex profiles and interaction logs for online information services run by the industry partner, SEEK. The new techniques are intended to redefine understanding of task-oriented search, and have the potential to reinvent the user experience for complex search tasks. This project will transform how practical search systems are measured within complex task scenarios. This will result in substantial economic impact by enabling businesses providing task-based search services to provide more customized offerings. Within the target domain (job search), this greatly enhances a service highly relevant to Australia's productivity.Read moreRead less
Spoken conversational search: contextual interactive techniques to support effective information search over a speech-only communication channel. This project will develop new techniques for effective information search using speech only, supporting improved information access for visually impaired people or in situations that require focused visual attention (e.g. driving). The techniques are based on a conversational approach to information search and presentation of results.
Talking about place: tapping human knowledge to enrich national spatial data sets. Place descriptions are a common way for people to describe a location, but no current tools are smart enough to understand them. Emergency call centres are risking lives, users of navigation or web services are frustrated and addressing these problems costs billions of dollars per year. This project comes with a novel, interdisciplinary approach to automatically interpret human place descriptions and will develop ....Talking about place: tapping human knowledge to enrich national spatial data sets. Place descriptions are a common way for people to describe a location, but no current tools are smart enough to understand them. Emergency call centres are risking lives, users of navigation or web services are frustrated and addressing these problems costs billions of dollars per year. This project comes with a novel, interdisciplinary approach to automatically interpret human place descriptions and will develop novel methods to capture placenames with their meaning for smarter databases and automatic interpretation procedures. This acquired knowledge will be an important step forward for Australia's data custodians and users. Australia's location information industry will gain a significant advantage on a highly competitive global market.Read moreRead less
Sub-collection retrieval: understanding and improving search engines. Search engines have become essential in many parts of daily life. This project aims to improve the accuracy of search engines, increasing the productivity of information seeking tasks and reducing frustration with poor search answers, thus improving the current gateway to the world of information.
Discovery Early Career Researcher Award - Grant ID: DE140100275
Funder
Australian Research Council
Funding Amount
$392,979.00
Summary
Beyond keyword search for ranked document retrieval. This project will develop novel approaches to efficient and effective ranked text retrieval using a new class of rank-aware algorithms derived from self-indexes. These algorithms can support complex statistical calculations on the fly. Efficient algorithm design for big data is an increasingly important problem as energy costs continue to soar and can now exceed hardware costs for big data consumers such as Google. In this project, two importa ....Beyond keyword search for ranked document retrieval. This project will develop novel approaches to efficient and effective ranked text retrieval using a new class of rank-aware algorithms derived from self-indexes. These algorithms can support complex statistical calculations on the fly. Efficient algorithm design for big data is an increasingly important problem as energy costs continue to soar and can now exceed hardware costs for big data consumers such as Google. In this project, two important problems in web search are explored: real-time indexing and long-form query answering. Using self-index algorithms, this project presents a road map to move beyond simple keyword-based ranked document retrieval, thus allowing us to efficiently meet more demanding information needs of users in the next decade.Read moreRead less
Data retrieval from massive information structures. Information search is an essential tool. But most current services regard the data as unstructured collections of independent documents, free of context. Next-generation search applications, such as over social networks, or corporate websites, or XML data sets, must account for the inherent relationships between data items, and must allow the efficient inclusion of search context. Queries should favour semantically local data, giving results th ....Data retrieval from massive information structures. Information search is an essential tool. But most current services regard the data as unstructured collections of independent documents, free of context. Next-generation search applications, such as over social networks, or corporate websites, or XML data sets, must account for the inherent relationships between data items, and must allow the efficient inclusion of search context. Queries should favour semantically local data, giving results that depend on the perceived state of the querier. This project will develop indexing and search techniques for massive structured data sets. The new search methods will incorporate theoretical advances and will be experimentally validated using industry-standard open-source distributed systems.Read moreRead less
TRIIBE TRacking Indoor Information BEhaviour. This project will research the passive tracking of user's mobile devices in indoor spaces correlating their spatial behaviour with their information needs to deliver personalised information. The project will create a system that enables owners of large buildings (for example, shopping malls, airports, universities) to better manage their spaces and services.
Natural language processing for automated validation of protein databases. The project aims to use natural language processing and information retrieval to reconcile and improve sources of biological information. Biological research has produced vast volumes of information about proteins, captured in structured resources (databases) and unstructured documents. However, the accuracy of much of this information is questionable. The project proposes to develop methods to validate data and reduce th ....Natural language processing for automated validation of protein databases. The project aims to use natural language processing and information retrieval to reconcile and improve sources of biological information. Biological research has produced vast volumes of information about proteins, captured in structured resources (databases) and unstructured documents. However, the accuracy of much of this information is questionable. The project proposes to develop methods to validate data and reduce the dramatic inconsistencies in protein information resources by leveraging observed correlations and complementarity between them, and specifically through targeted fact extraction from the biomedical literature. These methods will be applied at scale across millions of published articles, to infer and validate functional information.Read moreRead less