uznltk

uznltk — is a lightweight and convenient NLP (Natural Language Processing) library for the Uzbek language. It includes text cleaning, morphological analysis, number and text conversions, syllable splitting, and many other functions.

🔗 Links

• PyPI page
• GitHub page

👤 Authors

• Salaev Ulug'bek – ulugbek0302@gmail.com
• Omanov Jasur – jasuromonov77@gmail.com
• Zaripboyev Ollabergan – dewel000per@gmail.com

🔧 Install

pip install uznltk

🚀 Usage

from uznltk import *

📚 Functions

clean_text(text)

Corrects characters specific to the Uzbek language (g', o', ( ’ )).

clean_text("O'zbekistonda ta'lim kuchli rivojlanmoqda")
# Result: "O‘zbekistonda ta’lim kuchli rivojlanmoqda"

solid_sign(text)

Returns words with a ( ’ ) character as a list.

solid_sign("ta'lim bo'lishi oldindan ma'lum edi")
# Result: ['ta’lim', 'ma’lum']

lemmatize(text) and stem_word(text)

Extracts the stem of a word.

lemmatize("mexanizatorlashtirilganlardan")
# Result: "mexanizatorlashtirilgan"

number_to_text(number)

Converts a number to Uzbek text.

number_to_text(54)
# Result: "ellik to‘rt"

text_to_number(text)

Converts a number in text to numeric form.

text_to_number("yetmish olti")
# Result: 76

download(name)

Downloads various resources (e.g. books, news).

download("book")

clean_stopword(text)

Removes stop words from the text.

clean_stopword("salom dunyo, biz sen va u bilan bugun maktabga bordik")
# Result: "salom dunyo, bugun maktabga bordik"

syllables(text)

Divides words into syllables.

syllables("Bizga ma’lum ishlar yuz bermoqda!")
# Result: ['Biz-ga', 'ma’-lum', 'ish-lar', 'yuz', 'ber-moq-da!']

hyphenation(text)

Each word is divided into syllables and presented in a list.

hyphenation("salom dunyo")
# Result: ['sa-lom dunyo', 'salom dun-yo']

count_syllable(text)

Counts the number of syllables in the text.

count_syllable("Salom Dunyo")
# Result: 4

count_text(text)

Counts the number of words in the text.

count_text("Salom Dunyo")
# Result: 2

split_sentences(text)

Sorts the sentences in the text into lists.

split_sentences("Salom Dunyo. Bugun ob-havo qisman bulutli")
# Result: ['Salom Dunyo', 'Bugun ob-havo qisman bulutli']

split_words(text)

Extracts only words from the text (without IP, email, emoji, URLs) into a list.

split_words("sen 192.168.1.18 va helloworld@example.com elektron manzilidasan. Manba https://pypi.org")
# Result: ['sen', 'va', 'elektron', 'manzilidasan', 'Manba']

💡 Information

• The library is entirely designed for the Uzbek language.
• It includes basic NLP components such as number processing, lemmatization, and syntacticization.