TY - GEN
T1 - Text-to-SQL in the Wild
AU - Hazoom, Moshe
AU - Malik, Vibhor
AU - Bogin, Ben
N1 - Publisher Copyright:
© 2021 Association for Computational Linguistics.
PY - 2021
Y1 - 2021
N2 - Most available semantic parsing datasets, comprising of pairs of natural utterances and logical forms, were collected solely for the purpose of training and evaluation of natural language understanding systems. As a result, they do not contain any of the richness and variety of natural-occurring utterances, where humans ask about data they need or are curious about. In this work, we release SEDE, a dataset with 12, 023 pairs of utterances and SQL queries collected from real usage on the Stack Exchange website. We show that these pairs contain a variety of real-world challenges which were rarely reflected so far in any other semantic parsing dataset, propose an evaluation metric based on comparison of partial query clauses that is more suitable for real-world queries, and conduct experiments with strong baselines, showing a large gap between the performance on SEDE compared to other common datasets.
AB - Most available semantic parsing datasets, comprising of pairs of natural utterances and logical forms, were collected solely for the purpose of training and evaluation of natural language understanding systems. As a result, they do not contain any of the richness and variety of natural-occurring utterances, where humans ask about data they need or are curious about. In this work, we release SEDE, a dataset with 12, 023 pairs of utterances and SQL queries collected from real usage on the Stack Exchange website. We show that these pairs contain a variety of real-world challenges which were rarely reflected so far in any other semantic parsing dataset, propose an evaluation metric based on comparison of partial query clauses that is more suitable for real-world queries, and conduct experiments with strong baselines, showing a large gap between the performance on SEDE compared to other common datasets.
UR - http://www.scopus.com/inward/record.url?scp=85138378596&partnerID=8YFLogxK
M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???
AN - SCOPUS:85138378596
T3 - NLP4Prog 2021 - 1st Workshop on Natural Language Processing for Programming, Proceedings of the Workshop
SP - 77
EP - 87
BT - NLP4Prog 2021 - 1st Workshop on Natural Language Processing for Programming, Proceedings of the Workshop
A2 - Lachmy, Royi
A2 - Yao, Ziyu
A2 - Durrett, Greg
A2 - Gligoric, Milos
A2 - Li, Junyi Jessy
A2 - Mooney, Ray
A2 - Neubig, Graham
A2 - Su, Yu
A2 - Sun, Huan
A2 - Tsarfaty, Reut
PB - Association for Computational Linguistics (ACL)
Y2 - 6 August 2021
ER -