README.md

metadata

dataset_info:
  features:
    - name: text
      dtype: string
    - name: label
      dtype:
        class_label:
          names:
            '0': Avto
            '1': Ayollar
            '2': Dunyo
            '3': Foto
            '4': Iqtisodiyot
            '5': Jamiyat
            '6': Jinoyat
            '7': Madaniyat
            '8': O‘zbekiston
            '9': Pazandachilik
            '10': Qonunchilik
            '11': Salomatlik
            '12': Siyosat
            '13': Sport
            '14': Texnologiya
  splits:
    - name: train
      num_bytes: 892446788
      num_examples: 410200
    - name: validation
      num_bytes: 111174020
      num_examples: 51275
    - name: test
      num_bytes: 111663893
      num_examples: 51275
  download_size: 593012664
  dataset_size: 1115284701
task_categories:
  - text-classification
  - fill-mask
  - text-generation
language:
  - uz
tags:
  - uz
  - news
pretty_name: UzbekTextClassification
size_categories:
  - 100K<n<1M

Dataset Card for "uzbek_news"

Table of Contents

• Dataset Description
  • Dataset Summary
  • Supported Tasks and Leaderboards
  • Languages
• Dataset Structure
  • Data Instances
  • Data Fields
  • Data Splits
• Dataset Creation
  • Curation Rationale
  • Source Data
  • Annotations
  • Personal and Sensitive Information
• Considerations for Using the Data
  • Social Impact of Dataset
  • Discussion of Biases
  • Other Known Limitations
• Additional Information
  • Dataset Curators
  • Licensing Information
  • Citation Information
  • Contributions

Dataset Description

• Homepage: https://github.com/elmurod1202/TextClassification
• Repository: More Information Needed
• Paper: https://arxiv.org/pdf/2302.14494
• Point of Contact: More Information Needed
• Size of downloaded dataset files: 593 MB
• Size of the generated dataset: 522 MB
• Total amount of disk used: 1115 MB

Dataset Summary

Multi-label text classification dataset for Uzbek language and some sourcode for analysis. This repository contains the code and dataset used for text classification analysis for the Uzbek language. The dataset consists text data from 9 Uzbek news websites and press portals that included news articles and press releases. These websites were selected to cover various categories such as politics, sports, entertainment, technology, and others. In total, we collected 512,750 articles with over 120 million words accross 15 distinct categories, which provides a large and diverse corpus for text classification. It is worth noting that all the text in the corpus is written in the Latin script.

Please refer to paper and GitHub repository for further details.

Disclaimer: The team releasing UzTextClassification did not write this model card. This is HuggingFace version of the dataset that is created for mainly easy to access usage. The original dataset files can be accessed and downloaded from https://doi.org/10.5281/zenodo.7677431

Dataset Structure

Data Instances

default

• Size of downloaded dataset files: 593 MB
• Size of the generated dataset: 522 MB
• Total amount of disk used: 1115 MB

An example of 'train' looks as follows.

{
    "label": 14,
    "text": "Samsung Galaxy S21 Ultra eng yaxshi kamerofonlar reytingida 17-o‘rinni egalladi DxOMark laboratoriyasi mutaxassislari Samsung Galaxy S21 Ultra’ning asosiy ..."
}

Data Fields

The data fields are the same among all splits.

default

• text: a string feature.
• label: a classification label, with possible values including 'Avto' (0), 'Ayollar' (1), 'Dunyo' (2), 'Foto' (3), 'Iqtisodiyot' (4), 'Jamiyat' (5), 'Jinoyat' (6), 'Madaniyat' (7), 'O‘zbekiston' (8), 'Pazandachilik' (9), 'Qonunchilik' (10), 'Salomatlik' (11), 'Siyosat' (12), 'Sport' (13), 'Texnologiya' (14).

Data Splits

name	train	validation	test
default	410200	51275	51275

Citation Information

@proceedings{kuriyozov_elmurod_2023_7677431,
  title        = {{Text classification dataset and analysis for Uzbek 
                   language}},
  year         = 2023,
  publisher    = {Zenodo},
  month        = feb,
  doi          = {10.5281/zenodo.7677431},
  url          = {https://doi.org/10.5281/zenodo.7677431}
}

Contact

For any questions or issues related to the dataset or code, please contact [[email protected], [email protected]].