Release Bashkir Frequency Index v8.1 (105M tokens, 858k wordforms)

.gitattributes

@@ -1,60 +1,2 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.avro filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.lz4 filter=lfs diff=lfs merge=lfs -text
-*.mds filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
-# Audio files - uncompressed
-*.pcm filter=lfs diff=lfs merge=lfs -text
-*.sam filter=lfs diff=lfs merge=lfs -text
-*.raw filter=lfs diff=lfs merge=lfs -text
-# Audio files - compressed
-*.aac filter=lfs diff=lfs merge=lfs -text
-*.flac filter=lfs diff=lfs merge=lfs -text
-*.mp3 filter=lfs diff=lfs merge=lfs -text
-*.ogg filter=lfs diff=lfs merge=lfs -text
-*.wav filter=lfs diff=lfs merge=lfs -text
-# Image files - uncompressed
-*.bmp filter=lfs diff=lfs merge=lfs -text
-*.gif filter=lfs diff=lfs merge=lfs -text
-*.png filter=lfs diff=lfs merge=lfs -text
-*.tiff filter=lfs diff=lfs merge=lfs -text
-# Image files - compressed
-*.jpg filter=lfs diff=lfs merge=lfs -text
-*.jpeg filter=lfs diff=lfs merge=lfs -text
-*.webp filter=lfs diff=lfs merge=lfs -text
-# Video files - compressed
-*.mp4 filter=lfs diff=lfs merge=lfs -text
-*.webm filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.json -text

README.md

@@ -2,7 +2,7 @@
 license: cc-by-4.0
 language:
 - ba
-pretty_name: Bashkir Frequency Index (105M tokens, 81 Sources Super Corpus)
+pretty_name: Bashkir Frequency Index (105M tokens)
 task_categories:
 - text-classification
 configs:
@@ -26,18 +26,18 @@ tags:
 
 # Bashkir Frequency Index v8.1
 
-Word-frequency index for the Bashkir language, built from **105.01 million tokens** of clean monolingual text across **81 diverse authentic sources** (republican newspapers, literary magazines, 28 district regional press editions, digital library Kitaptar, classical literature, folklore archives, speech transcripts, and the complete Bashkir Wikipedia universe). Russian language admixture and OCR scanning noise were removed via sentence-level filtering and Diamond Clean v2 pipeline.
+Word-frequency index for the Bashkir language, computed over **105.01 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) | **634,193** | count ≥ 3 | Production NLP applications, spellcheckers, Huzhle, Duolingo-style SRS, T9 |
-| **`full`** | **858,284** | count ≥ 2 | Comprehensive morphology studies, dialect research, linguistic analysis |
+| **`public`** (recommended) | **634,193** | count ≥ 3 | NLP applications, spellchecking, text processing, linguistic research |
+| **`full`** | **858,284** | count ≥ 2 | Comprehensive morphology studies, lexical research |
 
 Additionally provided in the repository:
-- [`top_coverage_50.json`](top_coverage_50.json): the minimal set of **1,730 word forms covering 50% of all tokens** across the language.
-- [`top_500.json`](top_500.json): clean top-500 core vocabulary list for quick mobile lookups.
+- [`top_coverage_50.json`](top_coverage_50.json): the minimal set of **1,730 word forms covering 50% of all tokens** in the corpus.
+- [`top_500.json`](top_500.json): top-500 core frequency vocabulary list.
 
 ## Schema
 
@@ -45,13 +45,13 @@ Additionally provided in the repository:
 |---|---|---|
 | `rank` | int64 | Position by descending frequency |
 | `word` | string | Word form, lowercase Cyrillic |
-| `count` | int64 | Absolute frequency count (total occurrences) |
+| `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 | Translation / Function | Count | FPM | Doc Frequency |
+| Rank | Word | Function / Translation | Count | FPM | Doc Frequency |
 |---:|---|---|---:|---:|---:|
 | 1 | **һәм** | conjunction "and" | 1,419,398 | 13,516.0 | 1,024,812 |
 | 2 | **менән** | postposition "with" | 974,143 | 9,276.1 | 759,204 |
@@ -68,18 +68,18 @@ Additionally provided in the repository:
 
 | Tokens Covered | Forms Needed | Note |
 |---:|---:|---|
-| **25%** | **197** | Ultra-core grammatical connectives and pronouns |
+| **25%** | **195** | Core grammatical connectives and pronouns |
 | **50%** | **1,730** | Essential high-frequency lexicon |
-| **80%** | **18,315** | Conversational and daily news proficiency |
-| **90%** | **55,287** | Advanced literary & administrative reading |
-| **95%** | **129,354** | Comprehensive vocabulary mastery |
+| **80%** | **22,416** | General conversational and media vocabulary |
+| **90%** | **78,531** | Advanced literary & specialized vocabulary |
+| **95%** | **195,042** | Comprehensive lexicon coverage |
 
-## Method and Cleanups
+## Method
 
-- **Tokenization:** Regex lowercase tokenization over Cyrillic alphabet including all 9 Bashkir-specific letters (`Ә Ғ Ҙ Ҡ Ң Ө Ҫ Ү Һ`).
-- **Base Corpus:** 6,876,733 clean deduplicated sentences (105.01M tokens) from 81 leaf publications and archives.
-- **Russian Function Words Filter:** Context-verified removal of Russian function words (quotes, names, headers) that do not have homographs in Bashkir.
-- **Length and Noise Filters:** Glued words (>24 characters) and isolated non-Bashkir single letters dropped (kept single `ә`).
+- **Tokenization:** Regex lowercase tokenization over Cyrillic script including all 9 Bashkir-specific letters (`Ә Ғ Ҙ Ҡ Ң Ө Ҫ Ү Һ`).
+- **Corpus Base:** 6.87M sentences (105.01M 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
 
@@ -89,8 +89,6 @@ from datasets import load_dataset
 # Load recommended public configuration
 dataset = load_dataset("failed09/bashkir-frequency-index", "public")
 df = dataset["train"].to_pandas()
-
-# Inspect top-10 words
 print(df.head(10))
 ```