Constructing an Interactive Natural Language Interface for Relational Databases

By Fei Li, Hosagrahar Visvesvaraya Jagadish. Proceedings of the VLDB Endowment, 2014

517 citations recorded by the source corpus.

Natural language has been the holy grail of query interface designers, but has generally been considered too hard to work with, except in limited specific circumstances. In this paper, we describe the architecture of an interactive natural language query interface for relational databases. Through a carefully limited interaction with the user, we are able to correctly interpret complex natural language queries, in a generic manner across a range of domains. By these means, a logically complex English language sentence is correctly translated into a SQL query, which may include aggregation, nesting, and various types of joins, among other things, and can be evaluated against an RDBMS. We have constructed a system, NaLIR (Natural Language Interface for Relational databases), embodying these ideas. Our experimental assessment, through user studies, demonstrates that NaLIR is good enough to be usable in practice: even naive users are able to specify quite complex ad-hoc queries.

Subfields: Artificial Intelligence, Computer Networks and Communications, Information Systems

View Paper: https://www.semanticscholar.org/paper/89de826b17bb3a8009c20b7573f85a184827e7fd

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Networks and Communications

In Information Systems

Citations in the corpus

Prefix-to-SQL: Text-to-SQL Generation from Incomplete User Questions (2021, methodology): “Benchmarks such as Academic (Li and Jagadish, 2014), IMDB and Yelp (Yaghmazadeh et al.”
TypeSQL: Knowledge-Based Type-Aware Neural Text-to-SQL Generation (2018, methodology): “The methods of the Database community (Li and Jagadish, 2014; Yaghmazadeh et al., 2017) involve more hand feature engineering and user interactions with the systems.”
Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task (2018, background): “Data Splits For the ﬁnal training dataset, we also select and include 752 queries and 1659 questions that follow our annotation protocol from six existing datasets: Restaurants, GeoQuery, Scholar, Academic, IMDB, and Yelp.”
Grammar-based Neural Text-to-SQL Generation (2019, background): “Other well studied datasets include Restaurants (Ana-Maria Popescu and Kautz, 2003), Academic (Li and Jagadish, 2014) and WikiSQL (Zhong et al., 2017).”
Dynamic Hybrid Relation Network for Cross-Domain Context-Dependent Semantic Parsing (2021, background): “…language sentence into a logical form, known as semantic parsing, is a fundamental problem in natural language processing (Zelle and Mooney 1996; Zettlemoyer and Collins 2005; Wong and Mooney 2007; Zettlemoyer and Collins 2007; Li and Jagadish 2014; Yaghmazadeh et al. 2017; Iyer et al. 2017).”
CodeS: Towards Building Open-source Language Models for Text-to-SQL (2024, methodology): “The training portion encompasses 7,000 manually annotated samples, supplemented by an additional 1,659 samples sourced from six previous text-to-SQL datasets: Restaurants [50, 63], GeoQuery [80], Scholar [30], Academic [35], IMDB, and Yelp [72].”
NL-EDIT: Correcting Semantic Parse Errors through Natural Language Interaction (2021, background): “Natural language to SQL: Natural language interfaces to databases have been an active field of study for many years (Woods et al., 1972; Warren and Pereira, 1982; Popescu et al., 2003; Li and Jagadish, 2014).”
Natural Language Interfaces with Fine-Grained User Interaction: A Case Study on Web APIs (2018, methodology): “To encourage participants to come up with their own formulation of the query text, we showed them the task description in another language (we used Chinese and recruited [20] or in a recorded voice message [19].”
Neural Databases (2020, background): “There has been significant work on translating queries posed in natural language into SQL queries on a database whose schema is known [4, 23, 58], with extensions to semi-structured data and knowledge bases [7, 30].”
Improving Text-to-SQL Evaluation Methodology (2018, background): “Recent work has explored incorporating user feedback to improve accuracy (Li and Jagadish, 2014).”
Natural language to SQL: Where are we today? (2020, methodology): “For example, the studies in the natural language community [22, 44, 53, 46, 21, 24] exclude comparisons with [25, 35] from the database community.”
Towards Complex Text-to-SQL in Cross-Domain Database with Intermediate Representation (2019, background): “Li and Jagadish (2014) propose a query tree in their NLIDB system to represent the meaning of a question and it mainly serves as an interaction medium between users and their system.”
RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parsers (2019, background): “…(questions and SQL queries, with the accompanying schemas), including 1,659 examples lifted from the Restaurants (Popescu et al., 2003; Tang and Mooney, 2000), GeoQuery (Zelle and Mooney, 1996), Scholar (Iyer et al., 2017), Academic (Li and Jagadish, 2014), Yelp and IMDB (Yaghmazadeh et al., 2017)”
Program Synthesis from Natural Language Using Recurrent Neural Networks (2017, methodology): “Rule-based approaches have been developed to synthesize SQL queries [18], SmartPhone scripts [17], Java method specications [25], and Spreadsheet programs [9].”
SQLizer: query synthesis from natural language (2017, methodology): “As we show experimentally, Sqlizer achieves high precision across multiple different databases and significantly outperforms Nalir [Li and Jagadish 2014], a state-of-the-art system that won a best paper award at VLDB’14.”
Bridging the Semantic Gap with SQL Query Logs in Natural Language Interfaces to Databases (2019, methodology): “NLIDBs have various techniques of producing the metadata, whether through semantic parsing [41] or a designated lexicon of keywords [22].”
Building Natural Language Interfaces to Web APIs (2017, background): “Despite the extensive studies on NLI to other formal representations like relational databases [15], knowledge bases [4, 30], and web tables [18, 26], little research has been conducted on NLI to web APIs.”
SQLNet: Generating Structured Queries From Natural Language Without Reinforcement Learning (2017, methodology): “The study of translating natural language into SQL queries has a long history (Warren & Pereira, 1982; Androutsopoulos et al., 1993; 1995; Popescu et al., 2003; 2004; Li et al., 2006; Giordani & Moschitti, 2012; Zhang & Sun, 2013; Li & Jagadish, 2014; Wang et al., 2017).”
SParC: Cross-Domain Semantic Parsing in Context (2019, methodology): “studies in semantic parsing (Zettlemoyer and Collins, 2005; Artzi and Zettlemoyer, 2013; Berant and Liang, 2014; Li and Jagadish, 2014; Pasupat and Liang, 2015; Dong and Lapata, 2016; Iyer et al., 2017) were based on small and singledomain datasets such as ATIS (Hemphill et al.”
Model-based Interactive Semantic Parsing: A Unified Framework and A Text-to-SQL Case Study (2019, background): “To enhance parsing accuracy and user conﬁdence in practical applications, interactive semantic parsing has emerged as a promising solution (Li and Jagadish, 2014; He et al., 2016; Chaurasia and Mooney, 2017; Su et al., 2018; Gur et al., 2018; Yao et al., 2019).”