UzMorphAnalyser

https://pypi.org/project/UzMorphAnalyser
https://github.com/UlugbekSalaev/UzMorphAnalyser

UzMorpAnalyser tool is focused to make morphological analysis of Uzbek word based on morphemes. The tool includes Stemmer, Lemmatizer, Morphological Analyze methods. It is created as a python library and uploaded to PyPI. It is simply easy to use in your python project or other programming language projects via the API.

About project

The tool is focused to make morphological analysis of Uzbek word based on morphemes. The tool includes Stemmer, Lemmatizer, Morphological Analyze methods.

Quick links

• Github
• PyPI
• Web-UI

Demo

You can use web interface.

Features

• Stemmer
• Lemmatizer
• Lemmatizer with POS tag
• Extract Morphemes list
• Analyzer
• Analyzer with POS tag

Usage

Three options to run UzMorphAnalyser:

• pip
• API
• Web interface

pip installation

To install UzMorphAnalyser, simply run:

pip install UzMorphAnalyser

After installation, use in python like following:

# import the library
from UzMorphAnalyser import UzMorphAnalyser
# create an object 
analyzer = UzMorphAnalyser()
# call stem method
analyzer.stem('maktabimda')
# call lemmatize method
analyzer.lemmatize('maktabimda')
# call lemmatize method with POS tag
analyzer.lemmatize('maktabimda', analyzer.POS.NOUN)
# call analyze method
analyzer.analyze('maktabimda')
# call analyze method with POS tag
analyzer.analyze('maktabimda', analyzer.POS.NOUN)

API

API configurations:

• Method: GET

• Response type: string

• URL: https://nlp.urdu.uz:8080/uzmorphanalyser/stem

  • Parameters: word:string
  • Sample Request: https://nlp.urdu.uz:8080/uzmorphanalyser/stem?word=maktabimda

• https://nlp.urdu.uz:8080/uzmorphanalyser/lemmatize

  • Parameters: word:string, pos:string
  • Sample Request: https://nlp.urdu.uz:8080/uzmorphanalyser/lemmatize?word=maktabimda&pos=NOUN

• https://nlp.urdu.uz:8080/uzmorphanalyser/analyze

  • Parameters: word:string, pos:string
  • Sample Request: https://nlp.urdu.uz:8080/uzmorphanalyser/analyze?word=maktabimda&pos=NOUN

Note: argument pos is optional in all methods

Web-UI

The web interface created to use easily the library: You can use web interface here.

Options

When you use PyPI or API, you should use following options as POS tag of a word which is optional parameters of lemmatize() and analyze() metods:
NOUN Noun
VERB Verb
ADJ Adjective
NUM Numerical
PRN Pronoun
ADV Adverb

pos parameters is optional for lemmatize and analyze metods.

Result Explaining

It returns single word in a string type from each method, stem and lemmatize, that is stem and lemma of given word, respectively.

Result from analyze method

analyze method returns a response as list of dictionary which is may contain following keys:

 {'word', 'lemma', 'pos', 'affix','affixed','tense','person','cases','singular','plural','question','negative','impulsion','copula','verb_voice','verb_func'}: 

Documentation

See here.

Citation

@misc{ulugbek_salaev_2023_10155225,
  author       = {Ulugbek Salaev},
  title        = {{MODELING MORPHOLOGICAL ANALYSIS BASED ON WORD- 
                   ENDING FOR UZBEK LANGUAGE}},
  month        = nov,
  year         = 2023,
  publisher    = {Zenodo},
  doi          = {10.5281/zenodo.10155225},
  url          = {https://doi.org/10.5281/zenodo.10155225}
}

Contact

For help and feedback, please feel free to contact the author.