Part-of-Speech Tagging for Northern Kurdish

Manually tokenized and POS-tagged proverb in Northern Kurdish ‘As every tree stands over its roots, so does every human blossom with their mother tongue.’

This repository provides access to all trained and ready-to-use POS models and the training and test data discussed in our MWE-UD 2024 paper.

Repository structure

datasets

This folder contains the:

• gold_data.tsv the gold-standard manually annotated and tokenized dataset which consists of 136 sentences(2,937 tokens).
• kmr_mg-ud-complete-augmented.conllu the UD Kurmanji augmented dataset.

models

This folder contains 7 sub folders where you can find all our trained POS models. All sub folders contain two model variants original and augmented.

• Baseline
• HMM
• ExtraTrees
• AveragedPerceptron
• BiLSTM
• CRF
• NK-XLMR

The folders BiLSTM and NK-XLMR are empty since their models are very big and therefore uploaded to Google Drive.

BiLSTM model variants link: Google Drive (5.02 GB). Download the zip file and extract its content to BiLSTM sub folder.

NK-XLMR model variants link: Google Drive (2.46 GB). Download the zip file and extract its content to NK-XLMR sub folder.

pos_cli.py

This file offers a CLI interface to interact with the models. The file can be called within the command line with the following arguments:

• python pos_cli --pos_model CRF --training_data_type augmented --sentence "Leyla Qasim dixwest dengê kurdan li cîhanê bide bihîstin." --tokenization_method KLPT

pos_flask.py

An alternative interface to the cli one to interact all POS models. Rund this file and then navigate to http://127.0.0.1:5000/

The web interface to interact with all POS models.

Requirements

• Operating system: macOS / OS X · Linux · Windows
• Python version: Python 3.9.19

Clone the repo and run pip install -r requirements.txt to install all dependencies.

Cite this project

Please consider citing this paper, if you use any part of the data or the POS models:

@inproceedings{morad-etal-2024-part,
    title = "Part-of-Speech Tagging for {N}orthern {K}urdish",
    author = "Morad, Peshmerge  and
      Ahmadi, Sina  and
      Gatti, Lorenzo",
    editor = "Bhatia, Archna  and
      Bouma, Gosse  and
      Dogruoz, A. Seza  and
      Evang, Kilian  and
      Garcia, Marcos  and
      Giouli, Voula  and
      Han, Lifeng  and
      Nivre, Joakim  and
      Rademaker, Alexandre",
    booktitle = "Proceedings of the Joint Workshop on Multiword Expressions and Universal Dependencies (MWE-UD) @ LREC-COLING 2024",
    month = may,
    year = "2024",
    address = "Torino, Italia",
    publisher = "ELRA and ICCL",
    url = "https://aclanthology.org/2024.mwe-1.11",
    pages = "70--80",
    abstract = "In the growing domain of natural language processing, low-resourced languages like Northern Kurdish remain largely unexplored due to the lack of resources needed to be part of this growth. In particular, the tasks of part-of-speech tagging and tokenization for Northern Kurdish are still insufficiently addressed. In this study, we aim to bridge this gap by evaluating a range of statistical, neural, and fine-tuned-based models specifically tailored for Northern Kurdish. Leveraging limited but valuable datasets, including the Universal Dependency Kurmanji treebank and a novel manually annotated and tokenized gold-standard dataset consisting of 136 sentences (2,937 tokens). We evaluate several POS tagging models and report that the fine-tuned transformer-based model outperforms others, achieving an accuracy of 0.87 and a macro-averaged F1 score of 0.77. Data and models are publicly available under an open license at https://github.com/peshmerge/northern-kurdish-pos-tagging",
}

License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.