Zero-Shot Prompting Strategies for Table Question Answering with a Low-Resource Language
This work explores the application of zero-shot prompting strategies for table question answering (TQA) in Portuguese, focusing specifically on the Text2SQL task. This task involves translating questions posed in natural language into Structured Query Language (SQL) queries, which can be executed ag...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Ital Publication
2024-10-01
|
| Series: | Emerging Science Journal |
| Subjects: | |
| Online Access: | https://www.ijournalse.org/index.php/ESJ/article/view/2540 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850197302921134080 |
|---|---|
| author | Marcelo Jannuzzi Yuriy Perezhohin Fernando Peres Mauro Castelli Aleš Popovič |
| author_facet | Marcelo Jannuzzi Yuriy Perezhohin Fernando Peres Mauro Castelli Aleš Popovič |
| author_sort | Marcelo Jannuzzi |
| collection | DOAJ |
| description | This work explores the application of zero-shot prompting strategies for table question answering (TQA) in Portuguese, focusing specifically on the Text2SQL task. This task involves translating questions posed in natural language into Structured Query Language (SQL) queries, which can be executed against a database to answer the original question. Given the popularity of relational databases across various domains, advancements in this field can substantially impact the accessibility and democratization of data as simpler and more intuitive interfaces for database interaction are developed. Despite this significant potential, progress in developing Portuguese TQA solutions remains limited. The proposed approach leverages Large Language Models (LLMs)—specifically the GPT-3.5 and GPT-4 models—through zero-shot prompting. The primary objectives are to assess the effectiveness of such LLMs in this task and to identify the most suitable prompt styles. These are evaluated using a Portuguese translation of the popular Spider Text2SQL benchmark. Results reveal that the proposed approach can generate adequate SQL queries to answer Portuguese language questions about various databases, mainly when using GPT-4. The findings suggest that including schema information and database content in the prompts is critical for satisfactory outcomes.
Doi: 10.28991/ESJ-2024-08-05-020
Full Text: PDF |
| format | Article |
| id | doaj-art-f13f8bfcb89f41da8b04e12ed323c1f8 |
| institution | OA Journals |
| issn | 2610-9182 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | Ital Publication |
| record_format | Article |
| series | Emerging Science Journal |
| spelling | doaj-art-f13f8bfcb89f41da8b04e12ed323c1f82025-08-20T02:13:11ZengItal PublicationEmerging Science Journal2610-91822024-10-01852003202210.28991/ESJ-2024-08-05-020726Zero-Shot Prompting Strategies for Table Question Answering with a Low-Resource LanguageMarcelo Jannuzzi0Yuriy Perezhohin1Fernando Peres2Mauro Castelli3Aleš Popovič4NOVA Information Management School (NOVA IMS), Universidade NOVA de Lisboa, Campus de Campolide, 1070-312, Lisboa,NOVA Information Management School (NOVA IMS), Universidade NOVA de Lisboa, Campus de Campolide, 1070-312, Lisboa,NOVA Information Management School (NOVA IMS), Universidade NOVA de Lisboa, Campus de Campolide, 1070-312, Lisboa,NOVA Information Management School (NOVA IMS), Universidade NOVA de Lisboa, Campus de Campolide, 1070-312, Lisboa,Faculty of Computer and Information Science, University of Ljubljana, Ljubljana,This work explores the application of zero-shot prompting strategies for table question answering (TQA) in Portuguese, focusing specifically on the Text2SQL task. This task involves translating questions posed in natural language into Structured Query Language (SQL) queries, which can be executed against a database to answer the original question. Given the popularity of relational databases across various domains, advancements in this field can substantially impact the accessibility and democratization of data as simpler and more intuitive interfaces for database interaction are developed. Despite this significant potential, progress in developing Portuguese TQA solutions remains limited. The proposed approach leverages Large Language Models (LLMs)—specifically the GPT-3.5 and GPT-4 models—through zero-shot prompting. The primary objectives are to assess the effectiveness of such LLMs in this task and to identify the most suitable prompt styles. These are evaluated using a Portuguese translation of the popular Spider Text2SQL benchmark. Results reveal that the proposed approach can generate adequate SQL queries to answer Portuguese language questions about various databases, mainly when using GPT-4. The findings suggest that including schema information and database content in the prompts is critical for satisfactory outcomes. Doi: 10.28991/ESJ-2024-08-05-020 Full Text: PDFhttps://www.ijournalse.org/index.php/ESJ/article/view/2540text to sqlnatural language processingcomputational linguisticszero-shot prompting. |
| spellingShingle | Marcelo Jannuzzi Yuriy Perezhohin Fernando Peres Mauro Castelli Aleš Popovič Zero-Shot Prompting Strategies for Table Question Answering with a Low-Resource Language Emerging Science Journal text to sql natural language processing computational linguistics zero-shot prompting. |
| title | Zero-Shot Prompting Strategies for Table Question Answering with a Low-Resource Language |
| title_full | Zero-Shot Prompting Strategies for Table Question Answering with a Low-Resource Language |
| title_fullStr | Zero-Shot Prompting Strategies for Table Question Answering with a Low-Resource Language |
| title_full_unstemmed | Zero-Shot Prompting Strategies for Table Question Answering with a Low-Resource Language |
| title_short | Zero-Shot Prompting Strategies for Table Question Answering with a Low-Resource Language |
| title_sort | zero shot prompting strategies for table question answering with a low resource language |
| topic | text to sql natural language processing computational linguistics zero-shot prompting. |
| url | https://www.ijournalse.org/index.php/ESJ/article/view/2540 |
| work_keys_str_mv | AT marcelojannuzzi zeroshotpromptingstrategiesfortablequestionansweringwithalowresourcelanguage AT yuriyperezhohin zeroshotpromptingstrategiesfortablequestionansweringwithalowresourcelanguage AT fernandoperes zeroshotpromptingstrategiesfortablequestionansweringwithalowresourcelanguage AT maurocastelli zeroshotpromptingstrategiesfortablequestionansweringwithalowresourcelanguage AT alespopovic zeroshotpromptingstrategiesfortablequestionansweringwithalowresourcelanguage |