---
dataset_info:
  features:
  - name: name
    dtype: string
  splits:
  - name: train
    num_bytes: 207152
    num_examples: 8273
  download_size: 85115
  dataset_size: 207152
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
language:
- th
tags:
- name
- dataset
- OCR
- text-generation
- Thai
license: apache-2.0
---

# Thai Firstname Corpus

## Overview
The **Thai Firstname Corpus** is a collection of **8,273 unique Thai first names**, designed for various linguistic and technological applications. While comprehensive, this dataset may not cover all Thai names in existence.

## Features & Structure
- **Language:** Thai (th-TH)
- **Total Names:** 8,273
- **Format:** Single-field dataset containing only Thai first names
- **Field:**
  - `name` *(string)* – A Thai first name

## Potential Applications
This dataset can be utilized for:
- **Random name generation** – Suitable for applications, games, and simulations.
- **OCR training** – Improving Thai text recognition in images.
- **Form validation** – Ensuring accurate name input in databases.
- **Linguistic analysis** – Studying Thai naming patterns.
- **Testing Thai NLP systems** – Enhancing natural language processing models.

## Usage
To load the dataset, you can use the following code:

```python
from datasets import load_dataset

dataset = load_dataset("suchut/thai-firstname-corpus")
```

## Data Source & Collection
The names in this corpus are sourced from publicly available lists of common Thai first names. The dataset aims to represent widely used names across different regions of Thailand.

## Ethical Considerations
### Social Impact
This dataset contributes to the advancement of Thai language technology by improving:
- Optical Character Recognition (OCR)
- Name recognition in applications
- Software localization for Thai users

### Limitations & Biases
- The dataset may **not include all Thai names**, especially rare or newly coined ones.
- There could be **regional and generational biases** based on the source of the names.

### Privacy & Sensitivity
- The dataset contains **only first names**, with no personally identifiable information (PII).
- These names are **common and widely used**, reducing privacy concerns.