Streszczenie:
In this paper we propose a method for detecting neologisms in Bulgarian. The method combines several techniques: (i) preprocessing and organization of the data to facilitate efficient analysis; (ii) frequency analysis and extraction of new-word candidates; (iii) filtering, grouping and ranking of results. The method is tested on data from the Bulgarian National Corpus. The evaluation is based on qualitative rather than quantitative measures and is performed manually.
