navigation

OldSlavNet

OldSlavNet is a generic pre-modern Slavic dependency parser, described in:

• Pedrazzini, Nilo & Hanne M. Eckhoff. 2021. OldSlavNet: A scalable Early Slavic dependency parser trained on modern language data. Software Impacts 100063.

@article{oldslavnet,
  title={OldSlavNet: A scalable Early Slavic dependency parser trained on modern language data},
  author={Pedrazzini, Nilo and Eckhoff, Hanne M.}, 
  journal={Software Impacts},
  pages={100063},
  year={2021},
}

Version 3.0.0

Version 3.0.0 has simplified the parser’s configuration and usage.

• To install all the dependencies users only need to run a Makefile.
• To annotate a text or series of texts, they only need to add the pre-preocessed texts to a specific directory and run a single Shell script which by default will tag all the texts in that directory.
• A new Shell script has been added to train, give a name to and test your own model.