| [Conference Home Page] |  |
[Conference Tutorials Page] |
| [Conference Home Page] | |
[Conference Tutorials Page] |
A pre-conference tutorial
August 9, 1998
Université de Montréal
Montréal, Québec, Canada
| Isobel
Heald Département d'Anglais, Faculté des Lettres et Sciences Humaines, 11, Boulevard Lavoisier, 49045 Angers cedex 01, France. Tel. 241 35 21 21. Fax 241 35 21 19. heald@bule.univ-angers.fr. | Rémi
Zajac Computing Research Laboratory, New-Mexico State University, Box 30001 / Dept 3CRL, Las Cruces, NM 88003, USA. Tel. 505-646-5782. Fax. 505-646-6218. zajac@crl.nmsu.edu. |
There has been in recent years a surge of interest in Controlled Languages, as manifested by the creation of a series of workshops on Controlled Languages (CLAW). Controlled Languages are subsets of natural languages which have been restricted regarding their syntax and lexicon for the purpose of readability and/or ease of processing by machine, and especially for machine translation. A central issue for Controlled Languages is the elimination of ambiguity. Controlled Languages are used in technical documentation: user manuals or maintenance manuals for example.
The subject of Controlled Languages is of interest to NLP researchers and developers who wish to build NLP tools for the production and translation of technical documentation. It is also of interest to the industry and the technical writers and translators who wish to introduce the use of a Controlled Language in their practice.
No technical expertise is required for this tutorial except general knowledge about Technical Documentation and Natural Language Processing.
The tutorial will be divided into four parts:
We will first introduce Controlled Languages, give examples of Controlled Languages, how they are defined, by whom, and for what purpose. Concrete examples will use a particular controlled language, Simplified English (SE), as defined by the AECMA and used in the aeronautics industry. Examples will focus on the problems for which Controlled Languages have been defined and how these problems are solved.
The second part of the tutorial will focus on the topic of the definition of a Controlled Language for two of its essential components: syntax and the lexicon. We will illustrate problems that could arise in trying to restrict syntax; their consequences for the writer and the implications for NLP. For example, Controlled Languages rules can be inadequate or introduce unforeseen complexities. We will also outline a strategy for creating a controlled lexicon.
In the third part, we will look at the practical experience of the teaching of Simplified English to technical writers, the problems encountered, and the reactions of those concerned, and in particular, the reactions to the introduction of AECMA Simplified English to replace 'traditional' technical language.
Finally, we will give a survey of the NLP tools for creating, maintaining and using Controlled Languages. We will also mention specific problems encountered in processing Controlled Languages.