INTERNATIONAL WORKSHOP MuLTILINguAL RESOuRcES, TEcHNOLOgIES ANd EvALuATION fOR cENTRAL ANd EASTERN EuROPEAN LANguAgES

This paper discusses the building of the first Bulgarian– Polish–Lithuanian (for short, BG–PL–LT) experimental corpus. The BG–PL–LT corpus (currently under development only for research) contains more than 3 million words and comprises two corpora: parallel and comparable. The BG–PL– LT parallel corpus contains more than 1 million words. A small part of the parallel corpus comprises original texts in one of the three languages with translations in two others, and texts of official documents of the European Union available through the Internet. The texts (fiction) in other languages translated into Bulgarian, Polish, and Lithuanian form the main part of the parallel corpus. The comparable BG–PL–LT corpus includes: (1) texts in Bulgarian, Polish and Lithuanian with the text sizes being comparable across the three languages, mainly fiction, and (2) excerpts from E-media newspapers, distributed via Internet and with the same thematic content. Some of the texts have been annotated at paragraph level. This allows texts in all three languages and in pairs BG–PL, PL–LT, BG–LT, and vice versa to be aligned at paragraph level in order to produces aligned threeand bilingual corpora. The authors focused their attention on the morphosyntactic annotation of the parallel trilingual corpus, according to the Corpus Encoding Standard (CES). The tagsets for corpora annotation are briefly discussed from the point of view of possible unification in future. Some examples are presented.