Alexandru Timotin, Florin Teodor Tãnãsescu *
Structures for a Thesaurus of Technical Terminology
After his death and at the
request of the IEC, CER resumed the elaboration of
a thesaurus (1984), on the basis of the fund of terms defined
in IEV, setting up a Working Group - coordinated by Prof. Dr.
Eng. Florin Teodor Tãnãsescu - with the participation
of the Research Institute for Electrical Engineering of Bucharest
(ICPE) and of the Electrotechnical Chair of
the "Politehnica" University of Bucharest (PUB)
[5].
The Working Group developed the first French version of the IEC
Thesaurus, edited by the IEC Central Office in Geneva at the end
of 1986 [6].
This edition was presented as a structured alphabetic
list of terms, descriptors and synonyms, in a volume of 900 pages,
each descriptor being accompanied by its nearest semantic vicinity
and its references to the source publications.
The IEC proposed
the preparation of an English edition, containing the same concepts.
This led to a full revision of the information structures in the
form of a multilingual data bank of concepts. In 1989
a bilingual version, French and English, with 26,000 concepts
was ready [7].
The logical and lexical
structures were developed by Prof. Dr. Eng. Al. Timotin of PUB.
The information structures and the software were created by Dr.
Eng. S. Gheþaru of ICPE. Besides the concepts coming from
IEV, there were included concepts defined in other IEC publications,
as well as additional concepts, not defined but identified
through the context in all these publications and requested
to fill in the gaps of the obtained hierarchical structures.
Ever since, the moral and physical degradation of the informatics
equipment that had been used imposed the shifting on DOS-operating
computers and the thorough revision of the software, with many
difficulties referring to the acquirement of the necessary equipment
(PC network). The work of collecting the terms and concepts has
been continued, in order to include all the concepts defined until
1993, with some temporary technical means. A third language, Romanian,
has been added to the data bank, and the introduction of a file
of definitions in Romanian is in course of completion.
New informatics equipment became operational at the beginning
of 1995. The requirement for a transportable version led to a
complete revision of the data bank and of the software by Eng.
I. Lopãtan of PUB and Mathematician P. Onica of ICPE. Developed
due to the editing of new IEV chapters, in
1997 the thesaurus acquired funds have about 30,000 concepts
and about 33,000 terms in each of the three languages.
2. The thesaurus of technical concepts
2.1. Features of a multilingual thesaurus of concepts
The realization of a thesaurus of concepts for a natural language or for a
specific humanistic discipline would be an extremely difficult task,
as the conceptual range is very broad and the nuances of a term are
context-dependent, present vague differences, and develop in time.
On the contrary, the specific idiom of an exact or technical science
uses more clearly delimited concepts, created by specialists and
generally defined by them, a fact that facilitates their semantic analysis.
A thesaurus of technical concepts is an identification tool of a concept
within a limited conceptual fund. This aspect facilitates the
identification of the relations which link the concept to its closest
concepts, i.e. establishing its semantic vicinity. The semantic
vicinity can identify the concept in a more suggestive way - in terms of
its current usage - than a definition, which is often restrictive.
A thesaurus emphasizes sense correlation in a way that is invariant to
the change of presentation language.
A multilingual thesaurus of concepts can facilitate the monitoring of
the coherence of dictionaries, and allows their subsequent development.
It can be useful to authors and translators, especially when it is associated
with an internationally standardized terminology for texts of
scientific, commercial, and legal importance. A multilingual thesaurus
of concepts can efficiently support the computer-aided translation.
2.2. The data bank of a thesaurus of concepts
The multilingual thesaurus of concepts is a relational data bank
composed of files (Annex 1), each file having entries with the
same structure, containing particular information attached to the
data - concept, relation, term, document-source, etc.
The first file is the file of concepts, each concept
being represented in an abstract way in an entry by an alphanumeric code,
and certain taxonomic and logical characteristics. This file contains the
conceptual base of the thesaurus, a language-independent one.
The significance of the concepts is suggested by the semantic relations
between the considered concept and others from the same fund. These are
binary relations, hierarchical and nonhierarchical (associative), grouped
in a file of relations between concepts, language-independent too,
which describes the semantic structure of the thesaurus. The data bank of
a thesaurus also contains a file of sources, i.e. of the
language-independent characteristics, of the original documents
(type, edition, year, etc.).
106
