Versions
Description
Morfessor FlatCat is a method for learning of morphological segmentation of natural language. FlatCat uses a Hidden Markov Model with the states (morph categories) ``{prefix, stem, suffix, non-morpheme}`` for modeling the word internal ordering of morphs. The non-morpheme category (marked ZZZ) is intended to catch segments that are not proper morphs because they are fragments of a larger morph. FlatCat can be trained in an unsupervised or semi-supervised fashion.
Repository
https://github.com/aalto-speech/flatcat
Project Slug
morfessor-flatcat
Last Built
6 years, 8 months ago passed
Maintainers
Home Page
http://www.cis.hut.fi/projects/morpho/
Badge
Tags
hmm, morphology, morphotactics, segmentation, semi-supervised, unsupervised
Short URLs
morfessor-flatcat.readthedocs.io
morfessor-flatcat.rtfd.io
Default Version
latest
'latest' Version
master