vpermilp/lid-mhr

spellman LID corpus — mhr

118,885 rows / 40.8M chars of mhr-language text with provenance, built for training and evaluating the spellman language detector. Rows are single-line texts: news articles and library works (paragraph/sentence-group sized) plus Telegram/VK community posts and comments (wild register — emoji, hashtags, code-switching artefacts, folk spellings kept on purpose).

Row length: median 247 chars (p10 70, p90 674).

Configs

config	contents	rows
moderated (default)	news · library · wiki — edited, published sources	56,314
wild	tg-post · tg-comment · vk-post · vk-comment — Telegram and VK community posts and comments	62,571
all	union of both	118,885

from datasets import load_dataset
moderated = load_dataset("vpermilp/lid-mhr")            # default config
wild = load_dataset("vpermilp/lid-mhr", "wild")

Columns

column	meaning
text	the row (single line, whitespace-flattened)
lang	ISO 639-3 (mhr)
source	site store or Telegram channel handle
register	news · library · wiki · tg-post · tg-comment · vk-post · vk-comment
date	publication date when the source exposed one (YYYY-MM-DD)
id	source URL (crawl) or message/post id (Telegram, VK)

Registers

register	rows
news	56,314
vk-post	49,857
tg-post	12,073
tg-comment	641

Sources

source	register	rows
marielgazet-mhr	news	48,854
gazeta_mari_el	vk-post	19,867
kugarnya_official	vk-post	11,596
kugarnya-mhr	news	7,460
cholman	vk-post	7,237
yamde_lii	vk-post	6,015
kugarnya	tg-post	5,090
mariy_el_gazet	tg-post	4,948
mari_in_russia	vk-post	4,156
mari_kanal	tg-post	781
mary_muro	vk-post	619
mari_kanal	tg-comment	429
mari_yilme	tg-post	410
chonym_pochyn_murena	tg-post	378
mariushem	vk-post	367
MaripeopleofMoscow	tg-post	272
mariuver	tg-post	193
mari_yilme	tg-comment	109
chonym_pochyn_murena	tg-comment	80
mariy_el_gazet	tg-comment	11
MaripeopleofMoscow	tg-comment	8
mariuver	tg-comment	4
marimuro	tg-post	1

How it was built and cleaned

1. Orthographic gate — every row carries letters that never occur in Russian orthography (e.g. Sakha ө ү ҥ ҕ һ), with a per-language marker-density floor; Russian-function-word share gates code-switched rows, which are split into native sentence runs rather than dropped.
2. Structural cleaning — site chrome and per-channel boilerplate (lines repeated across ≥8 posts), self-promo tails, TV schedules, Telegram thread roots (auto-forwarded post copies) removed.
3. Structural repairs — soft hyphens removed and print-layout splits rejoined (hyphenated / lowercase-continued paragraphs merged in page order); site breadcrumbs, TV schedules, weather tables, chord sheets, sponsor tails and letter-spacing artefacts dropped; Latin h/x homoglyphs inside Cyrillic words normalized (127 rows dropped); long rows must carry marker letters in ≥2 tokens (3,301).
4. Content-free filter — emoji/URL/number-only rows, Latin-dominant or mojibake rows, HTML-entity leftovers dropped (157).
5. Cross-source dedup on a normalized key (casefold; URLs, handles, hashtags, punctuation, digits stripped) — catches site↔channel cross-posts and reposts (8,290 removed).
6. PII masking — phone numbers and e-mails replaced by <phone> / <email>; no author identities were ever stored.
7. Model judge — rows the current spellman model contradicts at ≥0.9 confidence outside the language's close-pair group are dropped (5).

No train/validation/test split is shipped: the spellman pipeline assigns splits content-addressed (crc32) at mix time.

License

The compilation, cleaning and annotations are released under the MIT license. Row texts are redistributed from the public sources listed above (newspapers, libraries, public Telegram channels and VK communities) for language-identification research; each row carries its source and id so upstream attribution is always recoverable.

Built 2026-08-30T12:23:29+00:00.