TaaS- Terminology as a Service
A new way to access terminology services was launched in June 2012. TaaS is an EU-funded project coordinated by Tilde, and developed together with the Cologne University of Applied Sciences, Kilgray, the University of Sheffield and TAUS. TaaS’ goal is to provide the industry (as well as the academia and society) with an important multilingual terminological resource in the most innovative IT platform (cloud).
The target user group of TaaS is language workers who, due to lack of time and resources, need a simplified process for preparing, storing, maintaining
, and sharing job-specific term glossaries which provides instant access to the translation equivalents of the terms. This project meets the needs of language workers, specialists , and companies; it gives them the opportunity to take an active part in the creation, maintenance and standardization of the most current terminological data. Hours of research could be saved if terminologists, translators, drafters, or specialists from different departments share their findings in one centralized place.
The TaaS platform will also provide APIs (Application Programming Interface) and exporting tools for sharing terminological data with major term banks. Based on the normal terminological workflow, the platform will be able to provide users with automatic extraction of monolingual term candidates from uploaded documents. For the extracted terms, automatic recognition of translation equivalents using different public and industry terminology databases would be provided. For terms not found in the term banks, equivalents would be automatically retrieved from different specialized aligned corpora. The clean-up process involving revision, editing
, or deleting would be automatic.
All services were developed taking into account national and international standards and the needs of users, as well as already existing services in translation tools, and machine translation and authoring systems. To be able to provide these services, the developers are using the state-of-the-art terminology extraction techniques (automatised linguistic and statistical corpora analysis) and term recognition algorithms.
The platform will provide access to updated terminology resulting in language specialists spending less time searching for terminology, thus ensuring more consistent translations. The services will be made available with the end of the project, mid 2014.
Article written by Matilda Soare, trainee at TermCoord