Graduated with a PhD in Deductive Databases from the University of Melbourne, taught there briefly, and have been an academic staff member at UNSW ever since. Have conducted research in information extraction (from tables and process descirptions), multimedia retrieval, distributed databases, and e-learning systems. Also involved in the development of two LMSs currently in use in CSE, online roleplay systems (MEPS and ORE, used both in UNSW and more widely), the development of the AIMS academic information management system (used across UNSW), a curriculum mapping system ... and generally trying to make effective use of the flood of useful data floating around in UNSW. Currently Deputy Head of School (Edu) in CSE. Supervisors: Ramamohanarao Kotagiri
The Carrick Institute is an initiative of the Australia federal government. It is aimed at genera... more The Carrick Institute is an initiative of the Australia federal government. It is aimed at generating strategic change in Australian University education, via grants and other awards to approximately $20 million annually. By previous Australian standards, the potential funding for projects is large. However, the Carrick Institute has a well focused vision, and grant applications need to be aligned with that vision. This paper first describes some key aspects of the Carrick vision, before describing four multi-institutional computing ...
… Conference on Storage and Retrieval for …, 1999
This paper describes a snapshot of work in progress on the development of an e cient le-access me... more This paper describes a snapshot of work in progress on the development of an e cient le-access method for similarity searching in high-dimensional vector spaces. This method has applications in, for example, image databases where images are accessed ...
High-dose chemotherapy followed by autologous stem cell transplantation (HD-ASCT) is now the acce... more High-dose chemotherapy followed by autologous stem cell transplantation (HD-ASCT) is now the accepted therapy for most patients with Hodgkin lymphoma (HL) that is not controlled by conventional multiagent chemotherapy and can result in long-term progression-free ...
Proceedings Eighth International Workshop on Research Issues in Data Engineering. Continuous-Media Databases and Applications, 2000
Abstract In this paper, we describe a prototype image retrieval sys-tem for distributed environme... more Abstract In this paper, we describe a prototype image retrieval sys-tem for distributed environments based on 2D Projection Interval Relationships (2D-PIR). The system transforms an image to a symbolic representation that describes spa-tial relationships among objects in the ...
The average number of disk accesses required to perform primary-key retrieval in systems using li... more The average number of disk accesses required to perform primary-key retrieval in systems using linear hasing can be reduced by employing the technique of partial expansions. This scheme maintains a more uniform distribution of records amongst the pages of the database file, thus reducing the average length of overflow chains and hence the average query cost. In this paper, we analyse the cost benefits of using linear hashing with partial expansions for partial-match retrieval and show that this type of indexing is effective for use with partial-match retrieval on highly dynamic files.
Proceedings of the 27th annual international conference on Research and development in information retrieval - SIGIR '04, 2004
This paper presents a new two-phase pattern (2PP) discovery technique for information extraction.... more This paper presents a new two-phase pattern (2PP) discovery technique for information extraction. 2PP consists of orthographic pattern discovery (OPD) and semantic pattern discovery (SPD) where the OPD determines the structural features from an identified region of a document and the SPD discovers a dominant semantic pattern for the region via inference, apposition and analogy. Then the discovered pattern is applied back into the region to extract required data items through pattern matching. We evaluated 2PP using 6500 data items and obtained effective result.
This paper describes of a computer-assisted peer assessment system (called "Peers") whi... more This paper describes of a computer-assisted peer assessment system (called "Peers") which is currently in use in marking assignments in a postgraduate database subject within the Master of Information Science course at the University of New South Wales. Students enrolled in the Advanced Database Management System subject are required to submit their major essay assignment on-line. In the week following
This paper is a brief survey of the current state of the technology used in supporting efficientc... more This paper is a brief survey of the current state of the technology used in supporting efficientcontent-based retrieval in index-based multimedia databases that use feature vectors to representthe multimedia data. One of the main obstacles in performing efficient content-based multimediaretrieval is the high-dimensionality of the feature vectors used to characterise the data. Previoussurveys on multidimensional access structures tried to cope
The Carrick Institute is an initiative of the Australia federal government. It is aimed at genera... more The Carrick Institute is an initiative of the Australia federal government. It is aimed at generating strategic change in Australian University education, via grants and other awards to approximately $20 million annually. By previous Australian standards, the potential funding for projects is large. However, the Carrick Institute has a well focused vision, and grant applications need to be aligned with that vision. This paper first describes some key aspects of the Carrick vision, before describing four multi-institutional computing ...
… Conference on Storage and Retrieval for …, 1999
This paper describes a snapshot of work in progress on the development of an e cient le-access me... more This paper describes a snapshot of work in progress on the development of an e cient le-access method for similarity searching in high-dimensional vector spaces. This method has applications in, for example, image databases where images are accessed ...
High-dose chemotherapy followed by autologous stem cell transplantation (HD-ASCT) is now the acce... more High-dose chemotherapy followed by autologous stem cell transplantation (HD-ASCT) is now the accepted therapy for most patients with Hodgkin lymphoma (HL) that is not controlled by conventional multiagent chemotherapy and can result in long-term progression-free ...
Proceedings Eighth International Workshop on Research Issues in Data Engineering. Continuous-Media Databases and Applications, 2000
Abstract In this paper, we describe a prototype image retrieval sys-tem for distributed environme... more Abstract In this paper, we describe a prototype image retrieval sys-tem for distributed environments based on 2D Projection Interval Relationships (2D-PIR). The system transforms an image to a symbolic representation that describes spa-tial relationships among objects in the ...
The average number of disk accesses required to perform primary-key retrieval in systems using li... more The average number of disk accesses required to perform primary-key retrieval in systems using linear hasing can be reduced by employing the technique of partial expansions. This scheme maintains a more uniform distribution of records amongst the pages of the database file, thus reducing the average length of overflow chains and hence the average query cost. In this paper, we analyse the cost benefits of using linear hashing with partial expansions for partial-match retrieval and show that this type of indexing is effective for use with partial-match retrieval on highly dynamic files.
Proceedings of the 27th annual international conference on Research and development in information retrieval - SIGIR '04, 2004
This paper presents a new two-phase pattern (2PP) discovery technique for information extraction.... more This paper presents a new two-phase pattern (2PP) discovery technique for information extraction. 2PP consists of orthographic pattern discovery (OPD) and semantic pattern discovery (SPD) where the OPD determines the structural features from an identified region of a document and the SPD discovers a dominant semantic pattern for the region via inference, apposition and analogy. Then the discovered pattern is applied back into the region to extract required data items through pattern matching. We evaluated 2PP using 6500 data items and obtained effective result.
This paper describes of a computer-assisted peer assessment system (called "Peers") whi... more This paper describes of a computer-assisted peer assessment system (called "Peers") which is currently in use in marking assignments in a postgraduate database subject within the Master of Information Science course at the University of New South Wales. Students enrolled in the Advanced Database Management System subject are required to submit their major essay assignment on-line. In the week following
This paper is a brief survey of the current state of the technology used in supporting efficientc... more This paper is a brief survey of the current state of the technology used in supporting efficientcontent-based retrieval in index-based multimedia databases that use feature vectors to representthe multimedia data. One of the main obstacles in performing efficient content-based multimediaretrieval is the high-dimensionality of the feature vectors used to characterise the data. Previoussurveys on multidimensional access structures tried to cope
Uploads