yusufbaykaloglu/University_Mevzuat_QA

University Mevzuat QA Dataset

Dataset Summary

The University Mevzuat QA Dataset is a Turkish question-answering dataset specifically designed for tasks involving university regulations. This dataset contains 13.8k rows of questions and answers extracted and curated from official university regulations. It is aimed at facilitating research and development in Natural Language Processing (NLP), particularly in Turkish QA systems, legal document analysis, and regulatory compliance automation.

Dataset Structure

### Columns

questions: A string containing a question related to university regulations. Questions cover a variety of topics such as remote education, student enrollment, and proficiency requirements.

answers: A string containing the corresponding answer derived from university regulations. Answers provide concise and accurate responses to the questions.

### Languages

Turkish

Applications

This dataset is suitable for:

• —Fine-tuning language models for Turkish question-answering tasks.

• —Building legal document question-answering systems.

• —Automating FAQ generation for university regulations.

• —Enhancing retrieval-augmented generation (RAG) systems for legal texts.

Source Data

Articles on Regulation: Full text of the articles is available at: turkish-university-mevzuat(2024)

Dataset Statistics

Citation Information

@dataset{university_mevzuat_qa,
  author = {Yusuf Baykaloğlu},
  title = {University Mevzuat QA Dataset},
  year = {2025},
  url = {https://huggingface.co/datasets/yusufbaykaloglu/University_Mevzuat_QA}
}

Contributions and Support

For any questions or feedback, please contact:

• —Author: Yusuf Baykaloğlu
• —Email