The EAGLET concept for Spoken Language Terminology

Work Packages WP4, WP5 and WP6 of the EAGLES Phase II project, LE 3-4244 (Telematics Applications Programme - Sector D/12: Language Engineering) aim to produce a Supplement to the Handbook of Standards and Resources for Spoken Language Systems edited by Gibbon, Moore and Winski.

This report develops an approach to providing of terminology descriptions for work on spoken language systems. The enterprise of providing a standard terminology will be known as EAGLET (EAGLES Termbank). The following areas are developed:

1. Onomasiological and semasiological criteria for spoken language terminology
2. The Handbook of Standards and Resources for Spoken Language Systems as a text base for standard terminology
3. Terminological macromodel
4. Terminological hyperlexicon based on the macromodel
5. Terminological micromodel
6. Terminological relational database based on the micromodel
7. Standardisation issues
8. Glossary as a basis for semantic definitions
9. Index as a basis for concordance generation

An operational prototype of the EAGLET relational database is available at the EAGLES SLWG project-internal telecooperation site.

The work described here is intended for inclusion in the Supplement to the EAGLES Handbook of Standards and Resources for Spoken Language Systems.

Appendices containing bibliographical material on terminology practice, as well as glossary and term list material are included.

An important feature of the EAGLET concept is that the terminology resources are designed to be generated automatically from an underlying formal specification. The EAGLET DB prototype web site is currently generated automatically on this principle.

A number of restrictions on the scope of the current draft interim report must be noted:

1. The status of the report at the present is, as intended, as a draft interim deliverable. Revision, with scrutiny by experts in the field of terminology, is planned for 1998.
2. The report in its final version cannot cover all aspects of the field of Terminology, either in theory or in practice; neither can it cover all aspects of the terminological field of spoken language systems. The goal is to provide an outline of current possibilities and pointers to future efficient development in this area, possibly as a specific enterprise in its own right at some later date.
3. A pragmatic approach to terminology acquisition, based on computer-readable resources generated in previous work on the Handbook of Standards and Resources for Spoken Language Systems, is taken.
4. General issues of automatic and semi-automatic computational terminology acquisition in the context of Information Retrieval, for instance with subsumption relations over documents as in terminological logic based approaches, are not addressed.