Skip to content

NoerNova/ShanNLP

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

38 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ShanNLP: Shan Natural Language Processing

experimental project and self-research inspired by PythaiNLP

Current State

  • corpus dict word: 19904 words (60% corvered and need more to collected)

Word Tokenization method

  • maximal_matching
  • pythainlp (newmm)

TODO

  • mining more shan words, poem
  • experiment more method to tokenize
    • word tokenize
    • sentent tokenize
    • subword_tokenize
    • tokenize with deep learning
  • spelling check
  • pos tagging
  • translation
  • word_vector

USAGE

Install

Clone this Repo

# this project using pythainlp dependecy
# - Trie data structure
# - newmm (experimental)

pip install -r requirements.txt
# or pip install pythainlp

Install with pip

pip install git+https://github.com/NoerNova/ShanNLP

Tokenization

maximal_matching bruce-force

from shannlp import word_tokenize

# start measure execute time
# start = time.time()

# # Example usage
input_text = "တိူၵ်ႈသွၼ်လိၵ်ႈသင်ၶၸဝ်ႈ တီႈဝဵင်းမိူင်းၶၢၵ်ႇ တႄႇပိုတ်ႇသွၼ်ႁဵၼ်းလိၵ်ႈ ပဵၼ်ပွၵ်ႈၵမ်းႁႅၵ်း မီးသင်ၶၸဝ်ႈ မႃးႁဵၼ်း 56 တူၼ်။"

# default tokenizer engine="mm" (maximal_matching)
print(word_tokenize(input_text))

# end measure execute time
# end = time.time()
# print(end - start)

# output
# ['တိူၵ်ႈ', 'သွၼ်လိၵ်ႈ', 'သင်ၶ', 'ၸဝ်ႈ', ' ', 'တီႈ', 'ဝဵင်း', 'မိူင်းၶၢၵ်ႇ', ' ', 'တႄႇ', 'ပိုတ်ႇ', 'သွၼ်', 'ႁဵၼ်းလိၵ်ႈ', ' ', 'ပဵၼ်', 'ပွၵ်ႈ', 'ၵမ်း', 'ႁႅၵ်း', ' ', 'မီး', 'သင်ၶ', 'ၸဝ်ႈ', ' ', 'မႃး', 'ႁဵၼ်း', ' ', '56', ' ', 'တူၼ်', '။']
# 0.7220799922943115

pythainlp newmm

from shannlp import word_tokenize
import time

# start measure execute time
start = time.time()

# Example usage
input_text = "တိူၵ်ႈသွၼ်လိၵ်ႈသင်ၶၸဝ်ႈ တီႈဝဵင်းမိူင်းၶၢၵ်ႇ တႄႇပိုတ်ႇသွၼ်ႁဵၼ်းလိၵ်ႈ ပဵၼ်ပွၵ်ႈၵမ်းႁႅၵ်း မီးသင်ၶၸဝ်ႈ မႃးႁဵၼ်း 56 တူၼ်။"

print(word_tokenize(input_text, engine="newmm", keep_whitespace=False))

# end measure execute time
end = time.time()
print(end - start)

# output
# ['တိူၵ်ႈ', 'သွၼ်လိၵ်ႈ', 'သင်ၶ', 'ၸဝ်ႈ', 'တီႈ', 'ဝဵင်း', 'မိူင်းၶၢၵ်ႇ', 'တႄႇ', 'ပိုတ်ႇ', 'သွၼ်', 'ႁဵၼ်းလိၵ်ႈ', 'ပဵၼ်', 'ပွၵ်ႈ', 'ၵမ်း', 'ႁႅၵ်း', 'မီး', 'သင်ၶ', 'ၸဝ်ႈ', 'မႃး', 'ႁဵၼ်း', '56', 'တူၼ်', '။']
# 0.7088069915771484

Digit convert

from shannlp.util import digit_to_text

print(digit_to_text("မႂ်ႇသုင်ပီမႂ်ႇတႆး ႒႑႑႗ ၼီႈ"))

# output
# မႂ်ႇသုင်ပီမႂ်ႇတႆး သွင်ၼိုင်ႈၼိုင်ႈၸဵတ်း ၼီႈ

num_to_word

from shannlp.util import num_to_shanword

print(num_to_shanword(2117))
# output သွင်ႁဵင်ၼိုင်ႈပၢၵ်ႇသိပ်းၸဵတ်း

shanword_to_num

from shannlp.util import shanword_to_num

print(shanword_to_num("ထွၼ်ႁဵင်ၵဝ်ႈပၢၵ်ႇၵဝ်ႈသိပ်းဢဵတ်း"))
# output -1991

text_to_num

from shannlp.util import text_to_num

print(text_to_num("သွင်ႁဵင်ၼိုင်ႈပၢၵ်ႇသိပ်းၸဵတ်းပီပူၼ်ႉမႃး"))
# output ['2117', 'ပီ', 'ပူၼ်ႉ', 'မႃး']

Date converter

need more reference for years converter

current reference
# https://shn.wikipedia.org/wiki/ဝၼ်းၸဵတ်းဝၼ်း_ၽၢႆႇတႆး

# MO: ပီတႆး 2117
# GA: ပီၵေႃးၸႃႇ 1385
# BE: ပီပုတ်ႉထ 2566
# AD: ပီဢိင်းၵရဵတ်ႈ 2023
from shannlp.util import shanword_to_date
import datetime

print(f"မိူဝ်ႈၼႆႉ: {datetime.date.today()}")
print(f"မိူဝ်ႈဝၼ်းသိုၼ်း {shanword_to_date('မိူဝ်ႈဝၼ်းသိုၼ်း')}")

# output
# မိူဝ်ႈၼႆႉ: 2023-06-15
# မိူဝ်ႈဝၼ်းသိုၼ်း 2023-06-13 00:51:14.597118

years convert

from shannlp.util import convert_years

# ပီ AD -> ပီတႆး
print(convert_years(2023, "ad", "mo"))
# output 2117

# ပီတႆး -> ပီပုတ်ႉထ
print(convert_years(2117, "mo", "be"))
# output 2566

# ပီပုတ်ႉထ -> ပီၵေႃးၸႃႇ
print(convert_years(2566, "be", "ga"))
# output 1385

Keyboard

from shannlp.util import eng_to_shn, shn_to_eng

print(eng_to_shn("rgfbokifcMj"))
# output မႂ်ႇသုင်ၶႃႈ

print(shn_to_eng("ေၺၺူၼ"))
# output apple

Citations

Wannaphong Phatthiyaphaibun, Korakot Chaovavanich, Charin Polpanumas, Arthit Suriyawongkul, Lalita Lowphansirikul, & Pattarawat Chormai. (2016, Jun 27). PyThaiNLP: Thai Natural Language Processing in Python. Zenodo. http://doi.org/10.5281/zenodo.3519354

BibText entry:

@misc{pythainlp,
    author       = {Wannaphong Phatthiyaphaibun and Korakot Chaovavanich and Charin Polpanumas and Arthit Suriyawongkul and Lalita Lowphansirikul and Pattarawat Chormai},
    title        = {{PyThaiNLP: Thai Natural Language Processing in Python}},
    month        = Jun,
    year         = 2016,
    doi          = {10.5281/zenodo.3519354},
    publisher    = {Zenodo},
    url          = {http://doi.org/10.5281/zenodo.3519354}
}