failed09/bashkir-frequency-index

Bashkir Frequency Index v8.3

Word-frequency index for the Bashkir language, computed over 105.96 million tokens of clean monolingual text across diverse public domains (periodicals, news media, literary publications, encyclopedic texts, books, and general web archives). Non-Bashkir language admixture and scanning artifacts were filtered using automated language-filtering pipelines.

Configurations

Config	Rows	Cutoff	Use Case
public (recommended)	643,903	count ≥ 3	NLP applications, spellchecking, text processing, linguistic research
full	872,146	count ≥ 2	Comprehensive morphology studies, lexical research

Additionally provided in the repository:

• top_coverage_50.json: the minimal set of 1,688 word forms covering 50% of all tokens in the corpus.
• top_500.json: top-500 core frequency vocabulary list.

Schema

Column	Type	Description
rank	int64	Position by descending frequency
word	string	Word form, lowercase Cyrillic
count	int64	Absolute frequency count (occurrences)
doc_freq	int64	Number of distinct sentences containing this word form
freq_per_million	double	Normalized instances per million tokens (FPM / IPM)

Sample (Top 10 Words)

Rank	Word	Function / Translation	Count	FPM	Doc Frequency
1	һәм	conjunction "and"	1,477,617	13,944.6	1,100,080
2	менән	postposition "with"	1,025,336	9,676.3	787,541
3	бер	numeral "one / a certain"	704,744	6,650.8	541,883
4	ла	particle "also / and"	577,765	5,452.5	435,535
5	ул	pronoun "he / she / it / that"	553,176	5,220.4	467,171
6	был	pronoun "this"	513,475	4,845.8	425,507
7	тип	postposition "for / sake of"	476,987	4,501.4	389,692
8	өсөн	quotative gerund "saying / as"	475,376	4,486.2	382,144
9	лә	particle "also / and"	355,751	3,357.3	276,484
10	буйынса	postposition/link "according to"	316,385	2,985.8	268,187

Token Coverage

Tokens Covered	Forms Needed
25%	191
50%	1,688
80%	17,971
90%	54,400
95%	127,746

Method

• Tokenization: Regex lowercase tokenization over Cyrillic script including all 9 Bashkir-specific letters (Ә Ғ Ҙ Ҡ Ң Ө Ҫ Ү Һ).
• Corpus Base: 7.21M sentences (105.96M tokens).
• Filtering: Automated function-word language filtering to remove foreign language leaks.
• Form Constraints: Isolated single-character forms removed (except ә); concatenated strings (>24 chars) removed.

Loading

from datasets import load_dataset

# Load recommended public configuration
dataset = load_dataset("failed09/bashkir-frequency-index", "public")
df = dataset["train"].to_pandas()
print(df.head(10))

Citation

@dataset{failed09_bashkir_frequency_index_2026,
  title     = {Bashkir Frequency Index v8.3},
  author    = {failed09},
  year      = {2026},
  publisher = {Hugging Face},
  url       = {https://huggingface.co/datasets/failed09/bashkir-frequency-index}
}