Exploring the Feasibility of Crowd-Powered Decomposition of Complex User Questions in Text-to-SQL Tasks
Natural Language Interfaces to Databases (NLIDB), also known as Text-to-SQL models, enable users with different levels of knowledge in Structured Query Language (SQL) to access relational databases without any programming effort. By translating natural languages into SQL query, not only do NLIDBs minimize the burden of memorizing the schema of databases and writing complex SQL queries, but they also allow non-experts to acquire information from databases in natural languages. However, existing NLIDBs largely fail to translate natural languages to SQL when they are complex, preventing them from being deployed in real-world scenarios and generalizing across unseen complex databases. In this paper, we explored the feasibility of decomposing complex user questions into multiple sub-questions — each with a reduced complexity — as a means to circumvent the problem of complex SQL generation. We investigated the feasibility of decomposing complex user questions in a manner that each sub-questi- doi
- 10.1145/3511095.3531282
- name
- Exploring the Feasibility of Crowd-Powered Decomposition of Complex User Questions in Text-to-SQL Tasks
- source
- supplied-acm-publisher-html
- acm_url
- https://dl.acm.org/doi/10.1145/3511095.3531282
- doi_url
- https://doi.org/10.1145/3511095.3531282
- license
- Copyright terms are stated on the ACM article page.
- summary
- Natural Language Interfaces to Databases (NLIDB), also known as Text-to-SQL models, enable users with different levels of knowledge in Structured Query Language (SQL) to access relational databases without any programming effort. By translating natural languages into SQL query, not only do NLIDBs minimize the burden of memorizing the schema of databases and writing complex SQL queries, but they also allow non-experts to acquire information from databases in natural languages. However, existing NLIDBs largely fail to translate natural languages to SQL when they are complex, preventing them from being deployed in real-world scenarios and generalizing across unseen complex databases. In this paper, we explored the feasibility of decomposing complex user questions into multiple sub-questions — each with a reduced complexity — as a means to circumvent the problem of complex SQL generation. We investigated the feasibility of decomposing complex user questions in a manner that each sub-questi
- source_pdf
- HT-2022_39_31_3511095/3511095.3531282.pdf
- import_kind
- full_text
- open_access
- false
- displayAuthor
- Sara Salimzadeh, Ujwal Gadiraju, Claudia Hauff, Arie van Deursen
- displayPublishTime
- 2022-06-28
- source_attribution
- Formatting converted from the supplied ACM publisher HTML and verified against the supplied ACM archival PDF.
- acm_source_attribution
- Converted from authorized ACM archival PDF; DOI 10.1145/3511095.3531282
Powered by Seed HypermediaOpen App