Automatic Processing of Natural-Language Electronic Texts with NooJ

In this article the first one-million corpus for the Belarusian NooJ module is represented. The given corpus has been built up of texts, patched up into sections by different subject categories. From the broad list of possible subject categories in the sections the corpus focuses on fiction, historic, medical, scien‐ tific, sociological literature, etc. Given a great number of similar subject catego‐ ries, the first one-million corpus can be considered as a first subject collection of texts for the Belarusian NooJ module. The text corpus is expected to be suitable for research in the following aspects: word polysemy processing of various texts, polysemic punctuation marks processing, and a new lexical items search. The first one-million corpus for the Belarusian NooJ module can be fully applicable in many fields of linguistic research.