Welcome to the HyperGeo server of the Artificial Intelligence and Information Analysis Laboratory at the Aristotle University of Thessaloniki.

Table of Contents

Home pages of HyperGeo partners

• MS&I : AEROSPATIALE MATRA - Matra Systèmes & Information, France
• NFESP : Nouvelles Frontières España, Spain
• UM : Telecomputing Engineering Group (GET), Department of Information Systems, Universidade do Minho, Portugal
  Hypergeo@UM
• IJS : Institut "Jozef Stefan", Laboratory for Open Systems and Networks, Slovenia
  Hypergeo@JSI
• FIM : Faculty of Informatics and Management, University of Education in Hradec Kralove, The Czech Republic
• CAS : CAS Software AG, Germany
  Hypergeo@CAS
• CU : Geographic Information Science Group, School of Informatics, City University, U.K.
  Hypergeo@CU
• CETIR : Centre Européen des Technologies de l'Information en milieu Rural, France
  Hypergeo@CETIR

[ top ]

Web pages related to HyperGeo project

• HYPERGEO home page
• HYPERGEO Info page (Reflector)
• HYPERGEO web page @ City University

[ top ]

AUT Contributions to HyperGeo Deliverables

• State of the art in natural language request analysis
• Natural language request analysis component (First draft)
• Natural language request analysis component (Updated document)

[ top ]

AUT Powerpoint Presentations at Reviews

• Summary of the research conducted during the first six-months.
  • Corpus acquisition
  • Corpus processing
  • Building word category maps
  • Baseline document retrieval system

[ top ]

Bibliographic References

• List of References on Statistical Language Processing. [ Long index.] [ Short index.]

[ top ]

(The tools for biography maintenance have been created by George Matas , matas@cmp.felk.cvut.cz)

Papers

• A. Georgakis and C. Kotropoulos and N. Bassiou and I. Pitas, Hypergeo:A data organization and retrieval system for tourist information, Proceedings of the IASTED Applied Informatics Int. Conf. (AI 2001) vol. 1, pp. 719-724, Innsbruck, Austria, 19-22 February 2001. Slides
• A. Georgakis and C. Kotropoulos and A. Xafopoulos and I. Pitas, MM-WEBSOM: A variant of WEBSOM based on order statistics, NSIP 2001, accepted for publication, Baltimore, USA, June 3-6, 2001.
• A. Georgakis and C. Kotropoulos and A. Xafopoulos and I. Pitas, Document organization and retrieval using SOMs and statistical language modeling, Wor. on Pattern Recognition in Information Systems (PRIS'01), accepted for publication, Setubal, Portugal, 6-7 July, 2001.
• N. Bassiou, C. Kotropoulos and I. Pitas, Hierarchical word clustering for relevance judgements in information retrieval, Wor. on Pattern Recognition in Information Systems (PRIS'01), accepted for publication, Setubal, Portugal, 6-7 July, 2001.

[ top ]

Programs

• AIIA contributions

  • Document Organization (code)
  • Document Organization (manual)
  • Word Class Map / Document Map Visualization (code)
  • Word Class Map / Document Map Visualization (manual)
  • Information Retrieval (code)
  • Information Retrieval (manual)
  • Corpus Processing & Feature Vector Extraction (code & data) (Last Updated: 16/01/01)
  • Corpus Processing & Feature Vector Extraction (brief manual & UTF) (Last Updated: 23/12/00)
  
  

• Various resources

  • DISCERN
  • MatLab SOM ToolBox
  • CMU-Cambridge Statistical Language Modeling Toolkit v2
  • Bow: A Toolkit for Statistical Language Modeling, Text Retrieval, Classification and Clustering
  • SMART (IR system)

Related Software Tools

• SGML: see Robin Cover's SGML Web Page
• NLP: see Software tools for NLP

[ top ]

HyperGeo Corpus & Other Corpora

• AIIA-HyperGeo Corpus based upon the following resources

  Tourist Information

  • Travel Guide Web Pages
  • European Touristic Information Database

  Data Distribution Institutions

  • European Language Resources Association (ELRA)
  • Linguistic Data Consortium (LDC)

  Available Corpora

  • The British National Corpus (BNC)
  • Corpora available from ELSNET
  • List of Corpora (Maintained by Dr Claire Warwick, OUCS/BNC)
  • International Computer Archive of Modern English (ICAME)

  Lexical Resources

  • WordNet® The original WordNet was developed by George A. Miller and his colleagues in Princeton. It is a computerized, multi-relation, monolingual English thesaurus which captures several relations such as hypernomy, antonymy, synonymy, holonymy and others between verbs, nouns, adjectives, and adverbs. It is freely available and may be used for any purposes, including commercial ones.
    • WordNet Home Page
    • The LSD/LDS/GermaNet project (German WordNet)

  Related Topics

  • Newspapers on the Web

[ top ]

Various Links

• Cartia: A company specializing in SOM-based applications
• CORBA Basics
• XML (Extensible Markup Language) 1.0 Specification
• WAP Knowledge (&Tools)

[ top ]