tugrulkaya/turkish-news-headlines

🇹🇷 Turkish News Headlines Dataset

Dataset Description

Turkish news headlines dataset for text classification tasks. Contains headlines from various news categories.

Dataset Summary

• Language: Turkish (tr)
• Task: Multi-class text classification
• Total Examples: 1,500
• Categories: 8
• License: CC-BY-4.0

Categories

The dataset contains 8 news categories:

Category	Count	Percentage
politika	300	20.0%
ekonomi	300	20.0%
spor	300	20.0%
teknoloji	300	20.0%
sağlık	300	20.0%
kültür-sanat	0	0.0%
dünya	0	0.0%
gündem	0	0.0%

Dataset Structure

Data Fields

• headline (string): News headline text
• category (string): News category
• date (string): Publication date (YYYY-MM-DD)
• source (string): Data source
• length (int): Character count
• word_count (int): Word count

Data Splits

Split	Examples
Train	1,050
Validation	225
Test	225

Usage

from datasets import load_dataset

# Load dataset
dataset = load_dataset("tugrulkaya/turkish-news-headlines")

# Access splits
train_data = dataset['train']
print(train_data[0])

# Example output:
# {
#   'headline': 'Yapay zeka teknolojileri hızla gelişiyor',
#   'category': 'teknoloji',
#   'date': '2024-06-15',
#   'source': 'manual_collection'
# }

Use Cases

✅ Text Classification

• News categorization
• Topic modeling
• Headline generation

✅ NLP Research

• Turkish language modeling
• Sentiment analysis baseline
• Transfer learning experiments

✅ Educational

• ML course projects
• NLP tutorials
• Classification examples

Dataset Creation

Source Data

Headlines collected from public sources and manually curated.

Annotation Process

• Manual category assignment
• Quality control checks
• Duplicate removal
• Format standardization

Data Quality

• ✅ Balanced across categories
• ✅ No duplicates
• ✅ Clean formatting
• ✅ Consistent labeling

Limitations

• Limited to Turkish language
• Focused on news domain
• Time period: 2023-2024
• May not represent all news sources

Citation

@misc{kaya2024turkish-news-headlines,
  author = {Tuğrul Kaya},
  title = {Turkish News Headlines Dataset},
  year = {2024},
  publisher = {Hugging Face},
  howpublished = {\url{https://huggingface.co/datasets/tugrulkaya/turkish-news-headlines}}
}

License

This dataset is licensed under CC-BY-4.0.

Contact

• Creator: Tuğrul Kaya
• Hugging Face: @tugrulkaya
• Issues: Report issues on the dataset page

Acknowledgments

• Turkish NLP community
• News sources (for public data)
• Hugging Face for hosting

---

Last Updated: December 2024