Vojko Gorjanc, Korpus FidaPLUS: nova generacija slovenskega referenčnega korpusa

The article presents the FidaPLUS corpus, an upgrade of the first Slovene reference corpusFida. FidaPLUS is an extensive collection of electronic texts which represents a balanced sample of heterogeneous contemporary texts in Slovene. In addition, the corpus is accompanied by a powerful, information-based concordancer, freely available for general use on the internet. The article focuses on the presentation of the improvements made to the new reference corpus, i.e. the improved lemmatization of corpus texts, new statistical methods for the collocator search, an upgrade of the concordancer interface, and the creation of an information network needed for the corpus analysis by the user. Information concerning the very structure of the corpus is also provided, since comprehension of the corpus composition is vital for the interpretation of language information. An attempt is also made to place the new corpus within the Slovene research area as a significant milestone not only for corpus studies, but for linguistics in general.