TUR2SQL: A Cross-Domain Turkish Dataset For Text-to-SQL

Ali Bugra Kanburoglu; F. Boray Tek

doi:10.1109/UBMK59864.2023.10286686

TUR2SQL: A Cross-Domain Turkish Dataset For Text-to-SQL

Ali Bugra Kanburoglu, F. Boray Tek

Department of Artificial Intelligence and Data Engineering

Isik University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

Abstract

The field of converting natural language into corresponding SQL queries using deep learning techniques has attracted significant attention in recent years. While existing Text-to-SQL datasets primarily focus on English and other languages such as Chinese, there is a lack of resources for the Turkish language. In this study, we introduce the first publicly available cross-domain Turkish Text-to-SQL dataset, named TUR2SQL. This dataset consists of 10,809 pairs of natural language statements and their corresponding SQL queries. We conducted experiments using SQLNet and ChatGPT on the TUR2SQL dataset. The experimental results show that SQLNet has limited performance and ChatGPT has superior performance on the dataset. We believe that TUR2SQL provides a foundation for further exploration and advancements in Turkish language-based Text-to-SQL research.

Original language	English
Title of host publication	UBMK 2023 - Proceedings
Subtitle of host publication	8th International Conference on Computer Science and Engineering
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	206-211
Number of pages	6
ISBN (Electronic)	9798350340815
DOIs	https://doi.org/10.1109/UBMK59864.2023.10286686
Publication status	Published - 2023
Event	8th International Conference on Computer Science and Engineering, UBMK 2023 - Burdur, Turkey Duration: 13 Sept 2023 → 15 Sept 2023

Publication series

Name	UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering

Conference

Conference	8th International Conference on Computer Science and Engineering, UBMK 2023
Country/Territory	Turkey
City	Burdur
Period	13/09/23 → 15/09/23

Bibliographical note

Publisher Copyright:
© 2023 IEEE.

Keywords

ChatGPT
Dataset
SQLNet
Text-to-SQL

Access to Document

10.1109/UBMK59864.2023.10286686

Cite this

Kanburoglu, A. B., & Tek, F. B. (2023). TUR2SQL: A Cross-Domain Turkish Dataset For Text-to-SQL. In UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering (pp. 206-211). (UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/UBMK59864.2023.10286686

@inproceedings{e7b8a6d70e3440528cb27e7badc0a9ee,

title = "TUR2SQL: A Cross-Domain Turkish Dataset For Text-to-SQL",

abstract = "The field of converting natural language into corresponding SQL queries using deep learning techniques has attracted significant attention in recent years. While existing Text-to-SQL datasets primarily focus on English and other languages such as Chinese, there is a lack of resources for the Turkish language. In this study, we introduce the first publicly available cross-domain Turkish Text-to-SQL dataset, named TUR2SQL. This dataset consists of 10,809 pairs of natural language statements and their corresponding SQL queries. We conducted experiments using SQLNet and ChatGPT on the TUR2SQL dataset. The experimental results show that SQLNet has limited performance and ChatGPT has superior performance on the dataset. We believe that TUR2SQL provides a foundation for further exploration and advancements in Turkish language-based Text-to-SQL research.",

keywords = "ChatGPT, Dataset, SQLNet, Text-to-SQL",

author = "Kanburoglu, {Ali Bugra} and Tek, {F. Boray}",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 8th International Conference on Computer Science and Engineering, UBMK 2023 ; Conference date: 13-09-2023 Through 15-09-2023",

year = "2023",

doi = "10.1109/UBMK59864.2023.10286686",

language = "English",

series = "UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "206--211",

booktitle = "UBMK 2023 - Proceedings",

address = "United States",

}

Kanburoglu, AB & Tek, FB 2023, TUR2SQL: A Cross-Domain Turkish Dataset For Text-to-SQL. in UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering. UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering, Institute of Electrical and Electronics Engineers Inc., pp. 206-211, 8th International Conference on Computer Science and Engineering, UBMK 2023, Burdur, Turkey, 13/09/23. https://doi.org/10.1109/UBMK59864.2023.10286686

TUR2SQL: A Cross-Domain Turkish Dataset For Text-to-SQL. / Kanburoglu, Ali Bugra; Tek, F. Boray.
UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering. Institute of Electrical and Electronics Engineers Inc., 2023. p. 206-211 (UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - TUR2SQL

T2 - 8th International Conference on Computer Science and Engineering, UBMK 2023

AU - Kanburoglu, Ali Bugra

AU - Tek, F. Boray

PY - 2023

Y1 - 2023

N2 - The field of converting natural language into corresponding SQL queries using deep learning techniques has attracted significant attention in recent years. While existing Text-to-SQL datasets primarily focus on English and other languages such as Chinese, there is a lack of resources for the Turkish language. In this study, we introduce the first publicly available cross-domain Turkish Text-to-SQL dataset, named TUR2SQL. This dataset consists of 10,809 pairs of natural language statements and their corresponding SQL queries. We conducted experiments using SQLNet and ChatGPT on the TUR2SQL dataset. The experimental results show that SQLNet has limited performance and ChatGPT has superior performance on the dataset. We believe that TUR2SQL provides a foundation for further exploration and advancements in Turkish language-based Text-to-SQL research.

AB - The field of converting natural language into corresponding SQL queries using deep learning techniques has attracted significant attention in recent years. While existing Text-to-SQL datasets primarily focus on English and other languages such as Chinese, there is a lack of resources for the Turkish language. In this study, we introduce the first publicly available cross-domain Turkish Text-to-SQL dataset, named TUR2SQL. This dataset consists of 10,809 pairs of natural language statements and their corresponding SQL queries. We conducted experiments using SQLNet and ChatGPT on the TUR2SQL dataset. The experimental results show that SQLNet has limited performance and ChatGPT has superior performance on the dataset. We believe that TUR2SQL provides a foundation for further exploration and advancements in Turkish language-based Text-to-SQL research.

KW - ChatGPT

KW - Dataset

KW - SQLNet

KW - Text-to-SQL

UR - http://www.scopus.com/inward/record.url?scp=85177574067&partnerID=8YFLogxK

U2 - 10.1109/UBMK59864.2023.10286686

DO - 10.1109/UBMK59864.2023.10286686

M3 - Conference contribution

AN - SCOPUS:85177574067

T3 - UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering

SP - 206

EP - 211

BT - UBMK 2023 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 13 September 2023 through 15 September 2023

ER -

Kanburoglu AB, Tek FB. TUR2SQL: A Cross-Domain Turkish Dataset For Text-to-SQL. In UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering. Institute of Electrical and Electronics Engineers Inc. 2023. p. 206-211. (UBMK 2023 - Proceedings: 8th International Conference on Computer Science and Engineering). doi: 10.1109/UBMK59864.2023.10286686

TUR2SQL: A Cross-Domain Turkish Dataset For Text-to-SQL

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this