Using Database Rule for Weak Supervised Text-to-SQL Generation

07/01/2019
by   Tong Guo, et al.
0

We present a simple and novel way to do the task of text-to-SQL problem with weak supervision. We call it Rule-SQL. Given the question and the answer from the database table without the SQL logic form, Rule-SQL use the database rules for the SQL exploration first and then use the explored SQL for supervised training. We design several rules for reducing the exploration search space. For the deep model, we leverage BERT for the representation layer and separate the model to SELECT, AGG and WHERE parts. The experiment result on WikiSQL outperforms the strong baseline of full supervision and is comparable to the start-of-the-art weak supervised mothods.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/24/2020

Structure-Grounded Pretraining for Text-to-SQL

Learning to capture text-table alignment is essential for table related ...
research
10/16/2019

Content Enhanced BERT-based Text-to-SQL Generation

We present a simple methods to leverage the table content for the BERT-b...
research
08/31/2018

Implementing WHERE and ORDER BY as spreadsheet formulas

The WHERE and ORDER BY clauses of the SQL SELECT statement select a subs...
research
01/07/2019

Popular SQL Server Database Encryption Choices

This article gives an overview of different database encryption choices ...
research
04/07/2021

Efficient and Accurate In-Database Machine Learning with SQL Code Generation in Python

Following an analysis of the advantages of SQL-based Machine Learning (M...
research
04/11/2020

In-Machine-Learning Database: Reimagining Deep Learning with Old-School SQL

In-database machine learning has been very popular, almost being a clich...
research
05/14/2023

QueryBooster: Improving SQL Performance Using Middleware Services for Human-Centered Query Rewriting

SQL query performance is critical in database applications, and query re...

Please sign up or login with your details

Forgot password? Click here to reset