Mission | Objectives | Plans | Topics | Projects | Services | Resources | Applications | Publications | Course | Links | People | Contracts | Contact

Mission

Objectives/Goals

• To promote the application of the highest standard and most up-to-date methods of Machine Learning techniques to the analysis of natural language text.
  
• To act as a consulting and service agent to the language industry in methods of data mining of natural language.
  
• To develop generic software to support a wide range of applications for processing text.
  
• To investigate and develop new methods of analysis of natural language.
  
• To foster relationships with other computational linguists around the world to allow for collaboration and sharing of resources and skills.
• To adapt and apply the methods of Minimum Message Length (MML) encoding to a range of computational linguistic problems.

Strategic Plans

• To collect together a repository of the best software available for the analysis of natural language
• To develop a generic system for the specification of semantic knowledge and apply it as markup of target texts.
• To review major topics in areas of our strengths and provide critical reviews, such as for unsupervised and supervised learning methods for POS tagging and word sense disambiguation.
• To develop approaches to Computational Linguistics that explicitly exploit the functionality of supercomputers.
• To develop a Workflow Methodology for the creation and automatic revision and regeneration of linguistic resources requiring both paper and multimedia publication.
• To provide a consulting service to researchers who wish to have data analysed by our methods.

back to top

Background Topics

Machine Learning:

• Supervised Learning
  
• Unsupervised Learning
  
• Association Rules
  
• Minimum Message Length Encoding (MML)
  

• XML Standards
  
• SGML Standards
  

• Application Tagsets
  
• Morphology
  
• Machine Readable Dictionaries
• Lexical Databases

back to top

Research Projects

• Data Mining of Semi-Structured Data
  
• Core Ontology of the Basque Language
  
• Core Ontology of the Basque Language (Some early results)
  
• Diachronic Phonology
  
• Intelligent Self-Learning Parser-editor for Dictionary Databases (Project Description)
• Electronic Reference materials for Language Learning and Translation - Proof of Concept
  
• Workflow Methodology for the Automatic Regeneration of Multi-media Linguistic Publications (Project Description)
• Software Architecture and Engineering for LT Systems
• Evidence Based Parsers
• Document Classification Workbench
• Named Entity Recognition
• The Computation of an Ontolexicon
• A Workbench for the markup of Systemic Functional Linguistics Linguistic
• Methods for Extracting Tacit Knowledge for Knowledge Management

Services

• Process Data with Data Mining Tools
  
• Develop Data Mining Tools
  
• Collaborate on Joint Development and Analysis
  

back to top

Resources

• Supervised Learning: Decision Trees and Decision Graph - DTREES & DGRAPH
  
• Unsupervised Learning: (Taxonomies) SNOB
  
• Modelling: Probabilistic Finite State Automata - PFSA
  

Storage:

• Web Database Management Systems
  

Processing:

• Supercomputer Processing
  

Language Processing:

• English
  
• Basque
  
• Chinese
  

Literature:

• A List of Books on Semantics and Discourse Processing

back to top

Applications

• Meassuring Relative Distance Between Natural Languages (Beijing and Cantonese Dialects)
  
• Effectiveness of Psychotherapy
  
• Part-of-Speech Tagging
  
• Semantic Tagging
  
• Systemic Functional Linguistic Analysis
  
• Computer-aided Qualitative Data Analysis
  
• Basque Multimedia Grammar and Dictionary
  
• Basque Dictionary Search Engine on the Web ( Currently, the site, containing 6 searchable Basque dictionaries, is still under construction and is password-protected for limited use. If you are interested in using the site for research purposes, please contact us at the addresses at the bottom of the page.)

Publications

• (2002) Patrick, J with Whitelaw, C. "Orthographic Tries in Language Independent Named Entity Recognition". Proceedings of CoNLL-2002, Taiwan,
• (2002) Patrick, J with Whitelaw, C. "Orthographic Tries in Language Independent Naemd Entity Recognition". Australasian Natural Lanaguge Processing Workshop, Canberra, Australia, pp 1-8, ISBN 1 86408 800 1.
• (2002) Patrick, J with Palko, D., Munro, R., & Zappavigna, M. "User Driven Example-Based Training for Creating Lexical Knowledgebases". Australasian Natural Lanaguge Processing Workshop, Canberra, Australia, pp 17-24, ISBN 1 86408 800 1.
• (2001). Patrick, J with Goyal, I., "Boosted Decision Graphs for NLPLearning Tasks". In: Proceedings of CoNLL-2001, Toulouse, France. http://lcg-www.uia.ac.be/conll2001/clauses/
• (2001). Patrick, J with S. Anthony, H. Qiao, C. Charles. "Computational semantic analysis for appraising change in language usage", (Eds.) H Bunt, I van der Sluis, E. Thijsse, 4th International Workshop on Computaional Semantics, IWCS-4. pp468-472.
• (2001). Patrick, J with P. Carter, & F. Deane, "EXCOVE and using videos in knowledge elicitation". International Journal of Human Computer Studies, 54 (3), 301-317.
• (2000). Patrick, J with J. Zhang & X. Artola-Zubillaga, "An Architecture and Query Language for a Federation of Heterogeneous Lexical and Dictionary Databases", Computers and the Humanities, 34: 393-407.
• (2000). Patrick, J with Balnaves, E. Kam, T. Ler, D. and Yeates, T. "A Case Study of Multimedia Architecture Design: Workflow and Version Control for Regeneration of Multimedia Systems". Grundy, J. & Han, J. (eds.). 3rd Australian Workshop on Software and System Architectures, pp88-99.
• (1999). Patrick, J with Azkue, J.J., Bidaurrazaga, A., "Gorputza, osasuna eta gaixoa Azkuerengan (I): gorputz atalen izendapena." Euskera, XLIV: 507-576.
• (1999). Patrick, J, "Tagging psychotherpeutic interviews for linguisitc analysis". Proc of Workshop on Towards Standards and Tools form Discourse Tagging. Assoc for Comp Linguisitics:New Brunswick. 58-64.
• (1998). Patrick, J with A.Raman & P. Andreae, "A Beam Search Algorithm for PFSA Inference", Pattern Analysis & Applications. 1:121-129. Springer-Verlag London Ltd.
• (2001). Patrick, J, A Student Grammar of Euskera, xv+pp400+xxv, Lincom Europa: Munich.
• (2001) Patrick, J with Hossain, L and Rashid, M.A. Enterprise Resource Planning: Opportunities and Challenges, Idea Group Publishing, ISBN:193070836X.
• (2002) Patrick, J with Palko, D., Munro, R., & Zappavigna, M. "Inferring Semantic Structure from Format". Digital Resources in the Humanities (ed.) C. Cole, Sydney, 2002 (in the press).
• (2001). Patrick, J with Hossain, L and Rashid, M.A. "The Evolution of ERP Systems: A Historical Perspective", in Enterprise Resource Planning: Opportunities and Challenges, (eds) Patrick, J., Hossain, L & Rashid, M.A. Idea Group Publishing.
• (2001). Patrick, J with Balnaves, E., Kam, T., Ler, D., Yeates, T., "A Case Study of Workflow and Version Control for Regeneration of Multimedia Systems", Computing Arts 2001: Digital Resources for Research in the Humanities, Uni. Sydney, http://setis.library.usyd.edu.au/drrh2001/
• (2001). Patrick, J with Balnaves, E. "Managing Multimedia Content Databases: toward a Model for Content Management", Computing Arts 2001: Digital Resources for Research in the Humanities, Uni. Sydney, http://setis.library.usyd.edu.au/drrh2001/
• (2001). Patrick, J with Palko, D. & Khan, A. "The Inductive Inference of Structure in Text Streams", Computing Arts 2001: Digital Resources for Research in the Humanities, Uni. Sydney, http://setis.library.usyd.edu.au/drrh2001/
• (2001). Patrick, J with Anthony, S. "Discourse Semantics for the Analysis of Change in Language", Computing Arts 2001: Digital Resources for Research in the Humanities, Uni. Sydney, http://setis.library.usyd.edu.au/drrh2001/
• (1998). Patrick, J with P. Carter and F.P. Deane, "Managing and analysing mulitple expert commentary of videotaped psychodrama sessions using FERAL", CIP98 Computers in Psychology, 15-17th April, York, UK.
• (1998) Patrick, J with E. Heinrich & E. Kemp, "A mulitmedia information system to support the analysis of human behaviour recorded on video", CIP98 Computers in Psychology, 15-17th April, York, UK.

back to top

Course

• Statistical Natural Language Processing

Links

• Acquilex Project
• GATE (general architecture for text engineering)
• Association for Computational Linguistics
• SIGNLL (ACL SIG for Natural Language Learning)
• SIGDAT (ACL SIG for Linguistic Data and Corpus-based Approaches to NLP)
• ELSNET
• Human Communication Research Centre, Edinburgh University
• Humanities Computing Unit, Oxford University
• Linguistic Data Consortium, Pennsylvania University
• "MIMIC"

back to top

People

• Stephen Anthony, PhD Scholar
  
• Hong Nguyen Pham, Ph.D. Scholar
  
• Prof. Jon Patrick, Director
  
• Casey Whitelaw, PhD Scholar
• Maria Herke Couchman, PhD Scholar
• Michele Zappavigna-Lee, PhD Scholar
• Ari Chanen, PhD Scholar
• Toby Hawker, PhD Scholar
• Denis Uy Dung Nguyen, Masters Scholar
• Andrea Stern, Senior Lecturer
• Matthew Honnibal, PhD Scholar
• Will Radford, Research Assistant
• John Drake, Honours Scholar
• Tim Yeates, Honours Scholar

Collaborators:

• Dr. David Dowe, Monash University
  
• A/Prof. David Powers, Flinders University
  
• Dr. Anand Raman, SRI
• Prof Jim Martin, Senior Researcher
• Prof Christian Matthiessen, Senior Researcher
• Dr Jane Simpson, Senior Researcher

Contracts

Capital Markets Co-operative Research Centre (CMCRC)
The development of a workbench for testing and calibrating document classification strategies including machine learners and feature selections.

Australian Securities and Investment Commission (ASIC)
The development of a document classification and search engine technology for identifying financial scams on the Internet.

back to top

Contact

Postal Address:

Visitor's Address:

Tel: +61 2 9351 4174
Fax: +61 2 9351 3838

back to top

back to School of Information Technologies page

back to main Sydney Uni page

Email webmaster. Last updated on September 8, 2005 .