---
license: apache-2.0
task_categories:
- text-generation
- text-classification
language:
- en
- tr
tags:
- turkish
- türkçe
- english
- ingilizce
- dataset
- general
- grade school
- primary school
- middle school
- secondary school
- high school
- university
- general culture
pretty_name: TR_EN Dataset
size_categories:
- 1M<n<10M
---

# Turkish–English General Dataset

**By Prometech AŞ**

🌐 [https://prometech.net.tr/](https://prometech.net.tr/)

📦 Total Size: **325 MB**

📄 License: **Apache License 2.0**

---

## 📌 Short Abstract

This dataset is a large-scale, multi-domain Turkish–English educational and technical corpus generated to support AI training, fine-tuning, and evaluation tasks. It combines K-12 education content, university-level Computer Engineering materials, and structured bilingual language learning data. The dataset is designed with high factual accuracy, curriculum alignment, and scalable generation in mind.

---

## 📚 Dataset Overview

The dataset is composed of **three main content domains**, generated programmatically and aggregated into a unified structure.

### 1. K-12 Education Data Generator

A function generates structured educational content for **Primary, Middle, and High School** levels.

**Subjects include:**

* Mathematics
* Science
* Social Studies

**Key Features:**

* Real-world curriculum-aligned topics
* Grade-appropriate explanations
* Structured schema for AI training
* Clear learning objectives and outcomes

---

### 2. Computer Engineering Data Generator (University Level)

A dedicated function produces **high-truth-value** Computer Engineering content suitable for undergraduate education.

**Covered Topics:**

* Algorithms
* Data Structures
* Operating Systems
* Software Engineering

**Key Features:**

* Technically accurate explanations
* Conceptual + applied knowledge
* Designed for LLM reasoning and instruction tuning

---

### 3. Turkish–English Language Training Data Generator

This function creates bilingual language learning pairs following **CEFR levels**:

* A1
* A2
* B1
* B2

**Focus Areas:**

* Grammar structures
* Vocabulary development
* Practical conversation examples
* Turkish ↔ English sentence pairs

---

## 🧩 Dataset Structure

All generated content is compiled into a **single Pandas DataFrame** with a consistent schema, including (but not limited to):

* `domain` (k12 / engineering / language)
* `level` (grade or CEFR level)
* `subject`
* `topic`
* `instruction`
* `input`
* `output`
* `truth_value` (for technical content)
* `language_pair` (for bilingual data)

---

## 📤 Export & Sample File

The generation script aggregates all data and exports it as:

* **JSON** or **CSV**

**Sample output path:**

```
/data/output/turkish_english_general_dataset_sample.json
```

This sample demonstrates the schema, formatting consistency, and content quality used as a seed toward the full dataset.

---

## 📈 Scaling to 325 MB (and Beyond)

To scale the dataset generation to the final **325 MB** volume:

1. Increase topic and curriculum coverage per domain
2. Expand grade-level and CEFR-level combinations
3. Use batch-based generation loops
4. Periodically append to the DataFrame
5. Export incrementally to avoid memory bottlenecks

The generation pipeline is fully scriptable and can be parallelized to reach larger dataset sizes efficiently.

---

## 🏢 About Prometech AŞ

**Prometech AŞ** is a technology company focused on AI, data engineering, and large-scale dataset creation for machine learning and language models.

🌐 [https://prometech.net.tr/](https://prometech.net.tr/)

---

## 📜 License

This dataset is released under the **Apache License 2.0**, allowing free use, modification, and distribution with proper attribution.