Skip to main content

Korean Inflectional Paradigm Generator

Project description

KoParadigm: Korean Inflectional Paradigm Generator

(Inflectional) paradigm means the set of all the inflected forms of a word. For example, English verb "look" has inflected forms like "look", "look-s", "look-ed", and "look-ing", as all of you know. Paradigms are widely used in corpus linguistics or search engines. To create the full paradigm set of a language is sometimes tricky. It is particularly so when we deal with a morphologically rich language like Korean. Inflection of Korean verbs is notorisouly complicated. Typically, a Korean verb can combine with more than 100 endings. What is worse, the combination rules are not simple at all. They are determined by the sound of the verb/ending, and the part-of-speech of the verb (action / descriptive). That's why so far there's no open sources of Korean paradigm generator, I think. Here's the first one. With KoParadigm, you can easily get the full paradigm of a Korean verb.

Dependencies

  • python >=3.6
  • jamo >=0.4.1
  • xlrd > 1.2.0

Installation

pip install koparadigm

Usage

>>> from koparadigm import Paradigm
>>> p = Paradigm()

>>> verb = "곱" # Note that you must drop the final ending 다
>>> paradigms = p(verb) # this returns list of lists
>>> print(paradigms)
[['동사', OrderedDict([('는다', '곱는다'), ('는다고', '곱는다고'), ('는다나', '곱는다나'), ('는다네', '곱는다네'), ('는다더라', '곱는다더라'), ('는다느니', '곱는다느니'), ('는다마는', '곱는다마는'), ('는다손', '곱는다손'), ('는담', '곱는담'), ('는답시고', '곱는답시고') ...
['형용사', OrderedDict([('습네', '곱습네'), ('습늰다', '곱습늰다'), ('습니까', '곱습니까'), ('습니다', '곱습니다'), ('습디까', '곱습디까'), ('습디다', '곱습디다'), ('습딘다', '곱습딘다'), ('습지요', '곱습지요'), ('으나', '고우나') ...]]
>>> for paradigm in paradigms:
...     print("pos =", paradigm[0])
...     for ending, form in paradigm[1].items():
...         print("ending =", ending, "form =", form)
...     print()
pos = 동사
ending = 는다 form = 곱는다
ending = 는다고 form = 곱는다고
ending = 는다나 form = 곱는다나
...

pos = 형용사
ending = 습네 form = 곱습네
ending = 습늰다 form = 곱습늰다
ending = 습니까 form = 곱습니까
...

References

If you use our software for research, please cite:

@misc{KoParadigm2019,
  author = {Park, Kyubyong },
  title = {KoParadigm},
  year = {2019},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/Kyubyong/paradigm}}
}

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

koparadigm-0.9.3.tar.gz (1.2 MB view hashes)

Uploaded Source

Built Distribution

koparadigm-0.9.3-py3-none-any.whl (1.2 MB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page