---
dataset_info: null
task_categories:
- text-generation
language:
- kk
- en
pretty_name: kaz-doctor-data
size_categories:
- 1K<n<10K
tags:
- medical
license: apache-2.0
---

# Dataset Card for "Kazakh-ChatDoctor-HealthCareMagic-10k"

[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)


# 🏥 ChatDoctor HealthCareMagic QA Dataset (Kazakh-English)

This dataset contains medical question-answer pairs collected and translated for the purpose of building and fine-tuning multilingual medical assistant models, especially for the Kazakh language.

## 📚 Dataset Overview

| Field          | Description                                                                 |
|----------------|-----------------------------------------------------------------------------|
| `instruction`  | Instruction to the model in English                                         |
| `input`        | User’s medical question or symptom description in English                   |
| `output`       | Doctor’s answer to the question in English                                  |
| `instruction_kk` | Instruction in Kazakh                                                       |
| `input_kk`     | Medical query or symptom description translated into Kazakh                 |
| `output_kk`    | Doctor’s response translated into Kazakh                                    |

The data follows the common instruction-tuning format (`instruction`, `input`, `output`) and is designed for fine-tuning language models in a medical Q&A context.

---

## ✨ Example

```CSV
{
  "instruction": "If you are a doctor, please answer the medical questions based on the patient's description.",
  "input": "I woke up this morning feeling the whole room is spinning...",
  "output": "Hi, Thank you for posting your query. The most likely cause is benign paroxysmal positional vertigo...",
  "instruction_kk": "Егер сіз дәрігер болсаңыз, науқастың сипаттамасына негізделген медициналық сұрақтарға жауап беріңіз.",
  "input_kk": "Мен бүгін таңертең ояндым, мен отырған кезде бүкіл бөлме айналып тұрғанын сездім...",
  "output_kk": "Сәлем, сұрауыңызды жібергеніңіз үшін рахмет. Сіздің симптомдарыңыздың ең ықтимал себептері – пароксизмальды позициялық вертиго..."
}
```

---

## 📊 Statistics

- **Number of samples**: ~5,541
- **Languages**: English, Kazakh
- **Use Case**: Instruction tuning, QLoRA, multilingual medical chatbots, low-resource model training

---

## ✅ Intended Use

- Fine-tuning medical LLMs (e.g., LLaMA, Mistral) with Kazakh medical data
- Research in multilingual and low-resource medical NLP
- Evaluation of machine translation quality in healthcare domain

---

## ⚠️ Disclaimer

This dataset is intended for research purposes only. It is **not a substitute for professional medical advice**, diagnosis, or treatment. Always consult a qualified healthcare provider with any medical questions.

---

## 📄 License

Apache 2.0

https://github.com/huggingface/datasets/blob/main/LICENSE