An Intelligent Part-Of-Speech tagger applied to Venda

The main objectives of this project are to try to make the use of a computer as natural as possible. A step towards achieving this is to develop a system that tags text and provides the user with an opportunity to correct tags and eliminate ambiguity of tags. From the data we now compute train data to be used on a Hidden Markov Model based tagger. We also compute word models and word frequencies that are used when improving the efficiency of the system. Venda text is applied to the system and this implies that we pay special attention to the unique diacritic of Venda.