Authors

Author1: Maksudbek

Author2: Dasturbek

Lemma & Lemmatization

The package finds lemmas of Uzbek words based on the dictionary.

The process of finding a lemma is called lemmatization.

There are 4 different ways of lemmatization: rule, dictionary, model, hybrid.

It is dictionary-based lemmatization algorithm [program, package].

Install & Clone

pip install UzbekLemma

git clone https://github.com/ddasturbek/UzbekLemma.git

Usage

import UzbekLemma as UL

print(UL.lemmatize("kelganlar")) #kelmoq

The algorithm flowchart

The dictionary structure

Scientific field

Patent

Some results of the program

Corpus & Results

We collected an equal number of texts from 23 different fields and stored them as a corpus.

We tested all the files (i.e. corpora) in the program and got these results.