Advances and challenges
Edited By Gabriel Quiroz Herrera and Pedro Patino Garcia
An Image-Term Co-occurrence Model for Multilingual Terminology Alignment and Cross-Language Image Indexing: Diego Burgos
An Image-Term Co-occurrence Model for Multilingual Terminology Alignment and Cross-Language Image Indexing
Nowadays, a great deal of specialized translation and terminology-based tasks must be carried out on the basis of rather static low-coverage textual terminological resources, e.g., specialized dictionaries, terminological databases, etc. Given the non-dynamic nature of most resources, some (or many) terms tend to become outdated and therefore do not reflect the conventional usage of terms among experts. Likewise, dictionaries and even terminological databases lag behind online and hardcopy product catalogues and technical manuals with regard to new terminology. As advances in science and new technologies take place, concepts change or emerge and terms show an evolution that is revealed too late in terminological resources.
Because of the above-mentioned low coverage and static conditions, dictionaries lack of the necessary information. This directly and indirectly affects the writing and translation of technical and scientific documentation. In the worst-case scenario, translators are not able to locate a suitable translation for a given term, in which case they come up with a new equivalent. However, as the search space for term translations is so big, it is probable that an equivalent already existed. If it did, the simultaneous and extensive practice of proposing new terms will likely result in an undetermined number of terminological vari ← 155 | 156 → ants throughout corpora and databases, and it would have consequences in specialized knowledge transference and discussion.
In an ideal setting, a concept...
You are not authenticated to view the full text of this chapter or article.
This site requires a subscription or purchase to access the full text of books or journals.
Do you have any questions? Contact us.Or login to access all content.