metadata
language:
- tg
- ru
- en
license: apache-2.0
tags:
- names
- tajik-names
- personal-names
- multilingual
Tajik Names Dataset
A multilingual dataset of Tajik personal names with parallel forms in Russian and English, including gender information.
📖 Description
This dataset contains 3,842 unique Tajik personal names, each accompanied by:
- Russian transliteration/translation
- English transliteration
- Gender (Male/Female)
It serves as a valuable resource for:
- Machine transliteration
- Gender classification from names
- Cross-lingual name matching
- Linguistic analysis of Tajik naming conventions
📊 Dataset Statistics
Overview
| Metric | Value |
|---|---|
| Total Records | 3,842 |
| Unique Tajik Names | 3,842 |
| Unique Russian Forms | 3,827 |
| Unique English Forms | 3,831 |
Gender Distribution
| Gender | Count |
|---|---|
| Male | 2243 |
| Female | 1599 |
| TOTAL | 3842 |
First Letter Distribution (Tajik Names)
| First Letter (Tajik) | Count |
|---|---|
| М | 543 |
| А | 356 |
| С | 338 |
| Ш | 274 |
| Н | 233 |
| Б | 217 |
| З | 170 |
| Г | 163 |
| Р | 153 |
| Ф | 149 |
| Ҳ | 143 |
| Д | 141 |
| Х | 129 |
| О | 111 |
| Ҷ | 90 |
| Т | 83 |
| В | 81 |
| К | 81 |
| П | 80 |
| И | 79 |
Name Length Statistics (characters)
| Language | Mean | Median | Min | Max |
|---|---|---|---|---|
| Tajik | 7.2 | 7 | 2 | 15 |
| Russian | 7.2 | 7 | 2 | 15 |
| English | 7.6 | 7 | 3 | 15 |
Most Frequent Tajik Names
| Tajik Name | Frequency |
|---|---|
| Аббос | 1 |
| Абдол | 1 |
| Абдуззоҳир | 1 |
| Абдулазиз | 1 |
| Абдулазим | 1 |
| Абдулалим | 1 |
| Абдулалӣ | 1 |
| Абдулаҳад | 1 |
| Абдулбасир | 1 |
| Абдулбашир | 1 |
📁 Data Format
Each record contains:
| Field | Type | Description |
|---|---|---|
Number |
int | Sequential ID (original order) |
Tajik |
string | Name in Tajik Cyrillic |
Russian |
string | Name in Russian Cyrillic |
English |
string | Name in Latin script |
Description |
string | Gender (Male/Female) |
🚀 Usage Examples
from datasets import load_dataset
# Load dataset
dataset = load_dataset("TajikNLPWorld/TajikNamesDataset")
# Get all male names
male_names = dataset["train"].filter(lambda x: x["Description"] == "Male")
# Find by first letter
names_starting_with_A = dataset["train"].filter(lambda x: x["Tajik"].startswith("А"))
# Iterate through records
for record in dataset["train"]:
print(record["Tajik"], "->", record["English"])
🔬 Potential Applications
- Transliteration models: Train on Tajik↔Russian↔English pairs.
- Gender classification: Build classifiers based on name strings.
- Cross-lingual entity linking: Match name variants across scripts.
- Linguistic analysis: Study common prefixes (Абд-, Абдул-) and suffixes.
📜 License
Apache 2.0
🤝 Citation
If you use this dataset, please cite:
@dataset{tajik_names_dataset_2026,
title = {Tajik Names Dataset},
author = {Arabov Mullosharaf Kurbonovich, TajikNLPWorld},
year = {2026},
publisher = {Hugging Face},
url = {https://huggingface.co/datasets/TajikNLPWorld/TajikNamesDataset}
}
Dataset automatically generated and published using Hugging Face 🤗 Datasets