chore: add initial dataset files

README.md

@@ -1,3 +1,97 @@
 ---
-license: mit
+language:
+- cv
+license: cc-by-sa-4.0
+tags:
+- chuvash
+pretty_name: "Chuvash Pre-Christian Names"
+size_categories:
+- 10k<n<100k
+annotations_creators:
+- expert-generated
+language_creators:
+- found
+source_datasets:
+- original
+task_categories:
+- other
+task_ids:
+- other
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/chuvash_names.jsonl
+dataset_info:
+  features:
+    - name: id
+      dtype: string
+    - name: name
+      dtype: string
+    - name: is_reconstructed
+      dtype: bool
+    - name: transcription
+      dtype: string
+    - name: transcription_latn
+      dtype: string
+    - name: gender
+      dtype: string
+    - name: variants
+      dtype: string
+    - name: etymology
+      sequence: string
+    - name: source
+      dtype: string
+  splits:
+    - name: train
+      num_examples: 12753
+  download_size: 4251417
+  dataset_size: 4251417
 ---
+
+# Chuvash Pre-Christian Names
+
+## Description
+
+This dataset is a structured version of **["Dictionary of Chuvash Anthroponyms of the Pre-Christian Era"](https://richfizh.chuvsu.ru/images/dokuments/Dictionary_%20Anthroponyms.pdf)** (A.M. Ivanova, E.V. Fomin, 2020). It contains over 10,000 given names used by the Chuvash people in the pre-Christian period (i.e., before the 19th century).
+
+## Usage
+
+The dataset can be loaded using the Hugging Face `datasets` library.
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("rustemgareev/chuvash-names", split='train')
+
+# Print the first example
+print(dataset[0])
+```
+**Example Output:**
+```json
+{
+  "id": "avakkay",
+  "name": "Аваккай",
+  "is_reconstructed": false,
+  "transcription": "авак:аj",
+  "transcription_latn": "avakkay",
+  "gender": "m",
+  "variants": null,
+  "etymology": ["араб. ‫أب اك ا‬ (Абака) ‘брат отца, друг’ + -ай"],
+  "source": "Ашмарин т. 1, с. 38"
+}
+```
+
+## Data Structure
+
+Each record is a JSON object with the following fields:
+
+*   `id` (string): A unique identifier in Latin script.
+*   `name` (string): The name in Cyrillic script.
+*   `is_reconstructed` (boolean): `true` if the name form is a linguistic reconstruction, `false` otherwise.
+*   `transcription` (string): Phonetic transcription in Cyrillic script.
+*   `transcription_latn` (string): Phonetic transcription in Latin script.
+*   `gender` (string): The gender associated with the name (`m` for male, `f` for female, `u` for unisex). Can be `null`.
+*   `variants` (string): Spelling variants of the name. Can be `null`.
+*   `etymology` (list of strings): A list containing one or more versions of the name's origin.
+*   `source` (string): A reference to the original source of information.