Improving Text-to-SQL Semantic Parsing with Fine-grained Query Understanding

TL;DR

This paper introduces a modular neural framework for Text-to-SQL parsing that uses fine-grained token-level understanding to improve accuracy and generalization, outperforming previous models on a new dataset.

Contribution

It proposes a novel token-level query understanding framework with NER, NEL, and semantic parsing modules, enhancing linkages between queries and databases.

Findings

Achieves 56.8% accuracy on WTQ dataset.

Outperforms state-of-the-art by 2.7%.

Demonstrates effectiveness of fine-grained understanding.

Abstract

Most recent research on Text-to-SQL semantic parsing relies on either parser itself or simple heuristic based approach to understand natural language query (NLQ). When synthesizing a SQL query, there is no explicit semantic information of NLQ available to the parser which leads to undesirable generalization performance. In addition, without lexical-level fine-grained query understanding, linking between query and database can only rely on fuzzy string match which leads to suboptimal performance in real applications. In view of this, in this paper we present a general-purpose, modular neural semantic parsing framework that is based on token-level fine-grained query understanding. Our framework consists of three modules: named entity recognizer (NER), neural entity linker (NEL) and neural semantic parser (NSP). By jointly modeling query and database, NER model analyzes user intents and identifies entities in the query. NEL model links typed entities to schema and cell values in database. Parser model leverages available semantic information and linking results and synthesizes tree-structured SQL queries based on dynamically generated grammar. Experiments on SQUALL, a newly released semantic parsing dataset, show that we can achieve 56.8% execution accuracy on WikiTableQuestions (WTQ) test set, which outperforms the state-of-the-art model by 2.7%.