Request Correction

Combining Concepts and Their Translations from Structured Dictionaries of Uralic Minority Languages

Building language resources for endangered languages, especially in the case of dictionaries, requires a substantial amount of manual work. This, however, is a time-consuming undertaking, and it is also why we propose an automated method for expanding the knowledge in the existing dictionaries. In this paper, we present an approach to automatically combine conceptually divided translations from multilingual dictionaries for small Uralic languages. This is done for the noun dictionaries of Skolt Sami, Erzya, Moksha and Komi- Zyrian in such a way that the combined translations are included in the dictionaries of each language and then evaluated by professional linguists fluent in these languages. Inclusion of the method as a part of the new crowdsourced MediaWiki based pipeline for editing the dictionaries is discussed. The method can be used there not only to expand the existing dictionaries but also to provide the editors with translations when they are adding a new lexical entry to the system.