THE BALANCED CORPUS OF MODERN LATVIAN AND THE TEXT SELECTION CRITERIA
Summary
Recently The Balanced Corpus of Modern Latvian (~3.5 million running words) has been created in the Institute of Mathematics and Computer Science (IMCS) (see http://www.korpuss.lv). The Corpus has been compiled from printed and electronic materials created after 1990. The Corpus is automatically morphologically tagged: for each token all the syntactically valid interpretations are stored.
Texts for the Corpus were chosen according to different text selection criteria: for instance, time, media, domain, etc. This article discusses the text selection criteria chosen for this Corpus, problems related to Corpus design and text selection criteria, solutions found for these problems and future plans regarding the Corpus.
This work is licensed under a Creative Commons Attribution 4.0 International License.