README.md

metadata

license: cc-by-sa-4.0
pretty_name: Russian Foreign Words
language:
  - ru
tags:
  - russian
  - dictionary
size_categories:
  - 10K<n<100K

Russian Foreign Words

This dataset is based on the Dictionary of Foreign Words developed by the Institute for Linguistic Studies of the Russian Academy of Sciences.

Usage

The dataset can be loaded using the Hugging Face datasets library.

from datasets import load_dataset

dataset = load_dataset("rustemgareev/russian-foreign-words", split='train')

Dataset Structure

Each entry in the dataset represents a dictionary article and is stored as a JSON object with the following fields:

• id (string): Unique identifier for the entry (e.g., "kafe"). Generated from the headword and optional homonym number (e.g., "kafedra_1").
• headword (string): The headword in lowercase, without stress marks. Used for easy searching and matching (e.g., "кафе").
• headword_stressed (string): The headword in lowercase with stress mark (U+0301). Preserves the original pronunciation guide (e.g., "кафе́").
• homonym_number (integer or null): Homonym index if the headword has multiple entries (e.g., 1, 2). null if there's only one entry.
• grammar (string): Raw grammatical information extracted from the source (e.g., "нескл., ср." for indeclinable neuter).
• etymology (string or null): Etymology of the word, typically enclosed in square brackets in the source (e.g., "франц. café < café кофе < араб. qahwa").
• source_language (string or null): The origin language extracted from the etymology field. Uses ISO 639-1 two-letter codes for most languages (e.g., "fr" for French, "ar" for Arabic); for Ancient Greek, the three-letter code "grc" (ISO 639-3) is used. null if language cannot be determined.
• see_also (string or null): Cross-reference to another entry. Present when the article is a redirect or suggests looking up another word (e.g., "СМ. АВИАЦИЯ").
• senses (array of objects): List of meanings (senses) for the headword. Each object contains the following fields:
  • number (integer or null): The sense number if multiple meanings exist. null for single-meaning entries or introductory text.
  • domain (string or null): Subject area or domain label (e.g., "В музыке" – "In music"). Applies to the specific sense.
  • style_label (string or null): Stylistic or register label (e.g., "Разговорное" – "Colloquial"). Note that abbreviations like "Разг." are expanded to full forms during processing.
  • definition (string): The definition or explanation of the sense.
  • examples (array of strings): Usage examples illustrating the sense in context. Each example is split into individual sentences.
  • collocations (array of objects): Common phrases or compound terms related to the sense. Each object has:
    • text (string): The phrase itself.
    • definition (string): Optional explanation of the phrase.

Example Entry

{
  "id": "kafe",
  "headword": "кафе",
  "headword_stressed": "кафе́",
  "homonym_number": null,
  "grammar": "нескл., ср.",
  "etymology": "франц. café < café кофе < араб. qahwa",
  "source_language": "fr",
  "see_also": null,
  "senses": [
    {
      "number": null,
      "domain": null,
      "style_label": null,
      "definition": "Предприятие общественного питания более низкого класса, чем ресторан, с подачей закусок, несложных горячих блюд, напитков, кондитерских изделий",
      "examples": [
        "Дорогое, дешевое кафе.",
        "Летнее кафе.",
        "Снять кафе для проведения банкета.",
        "Работать поваром, официантом в кафе.",
        "Пообедать в кафе."
      ],
      "collocations": []
    }
  ]
}

License

The dataset is distributed under the Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) license.