Automatic terminology checking

OData support
Dr. Asztalos Márk
Department of Automation and Applied Informatics

In the 21st century we live in a world, where a tiny invention of any small country could roam all over the world. For this, it is essential to pick up on it at every part of the world. This only can happen, if we speak a common language. This fact is an open secret for every great company and based on this, a new industry emerged, duty of which to translate documents, advertisements, movies and any other media to numerous language.

Due to the globalization, the translation procedure has to finish immediately. There are movie scripts, on which the ink has not been dried, but it is already sent to be synchronized. This process can only succeed cost-efficiently in time, if it is supported by modern software.

The aim of the project was to create a software, with which the translation process can be geared up and its correctness can be checked automatically. Thus it is necessary to handle terminological databases, based on which bilingual documents can be analysed. To reach this goal, I implemented my own search engine, which is adjusted to the needs of the translators. The search provides new capabilities and features, which are not included in the existing modern platforms, for example searching for partial or whole composition, or handling multiple termbase at the same time. On the basis of this engine, a new service is created to analyse written text and create statistics from its terminological correctness. It can handle both mono- and bilingual documents.

Based on this service a new application was designed to help the translators’ everyday life with its global hotkeys and quick feedbacks. Pages of monolingual text can be analysed in a few moments by only pressing a single button, depending on the selected terminological databases. Finally, a web application has been developed, which is capable of analysing bilingual documents.

The developed translation software products and its easily extensible API provides efficient possibilities for terminologist to do their best in work, thus spare time and money for their company.


Please sign in to download the files of this thesis.