nlidb-datasets
SENLIDB Dataset
This dataset contains 24,890 unique SQL queries, extracted from Stack Exchange Data Explorer site (http://data.stackexchange.com/stackoverflow/queries).
In addition to the information available on the original site, this dataset contains 780 additional manual annotations for 296 queries.
Contents:
-
README.md: this file
-
train.json:
Contains 24594 entries, with the following structure:
{
title : "(string) Title provided in natural language that explains the meaning of the SQL query"
description : "(string) A longer description of the query (may be missing)"
sql_plain : "(string) The original query, as it appeared on data.stackexchange"
sql : "(string) Query with comments removed"
comments : "(list of strings) List with the removed comments"
url : "(string) URL where the query or its revision can be found"
id : "(string) Unique identifier, extracted from the url"
}-
test.json:
Contains 296 entries that follow the structure from the training file, with the addition of the following fields:
{
annotations: [
{
annotation : "(string) Natural Language manual annotation"
annotator_id: "(integer) Number uniquely identifying the annotator"
}
]
}Acknowledgements
This work has been funded by the Text2NeuralQL research project (PN-III-P2-2.1-PTE-2016-0109).
License
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.
Reference
Brad, Florin, Radu Iacob, Ionel Hosu, and Traian Rebedea. "Dataset for a Neural Natural Language Interface for Databases (NNLIDB)." arXiv preprint arXiv:1707.03172 (2017). https://arxiv.org/abs/1707.03172