CMC training corpus Janes-Syn 1.0

Janes-Syn is a syntactically annotated corpus of Slovene tweets and is meant as a gold-standard training and testing dataset for syntactic annotation of Slovene computer-mediated communication and for detailed linguistic explorations which require highly accurate and reliable annotations. Words in the dataset are normalised, lemmatised, PoS-tagged and syntactically annotated with the JOS dependency model (http://eng.slovenscina.eu/tehnologije/razclenjevalnik). The annotations on all levels were manually corrected. The corpus creation and structure are described in: ARHAR HOLDT, Spela, FISER, Darja, ERJAVEC, Tomaž, KREK, Simon. Syntactic annotation of Slovene CMC : first steps. Proceedings of the 4th Conference on CMC and Social Media Corpora for the Humanities, 27-28 September 2016, Ljubljana, Slovenia, 2016, pp. 3-6. http://nl.ijs.si/janes/cmc-corpora2016/proceedings/ Janes-Syn was created from two larger corpora that are also available in the repository: Janes-Norm (http://hdl.handle.net/11356/1084) and Janes-Tag (http://hdl.handle.net/11356/1123).