The tagger parameter file is trained on an excerpt of the Pretoria Sepedi Corpus (D. Prinsloo, University of Pretoria): Here, about 5000 tokens were manually tagged and used for training the RF-Tagger (Helmut Schmid and Florian Laws: Estimation of Conditional Probabilities with Decision Trees and an Application to Fine-Grained POS Tagging, COLING 2008, Manchester, Great Britain). The tagger is freely available for academic purposes (see http://www.cis.uni-muenchen.de/~schmid/tools/RFTagger/). Methods and validation results can be found in: G. Faaß, U. Heid, E. Taljard, and D.J. Prinsloo. Part-of-Speech tagging in Northern Sotho: disambiguating polysemous function words. In Proceedings of the 1st Workshop on Language Technologies for African Languages - AfLaT 2009 at EACL, pages 38-45, Athens, Greece, 2009.