A Natural Language Interface for Database: Achieving Transfer-learnability Using Adversarial Method for Question Understanding
Authors:
- 1 April 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 97-108
Abstract
Relational database management systems (RDBMSs) are powerful because they are able to optimize and execute queries against relational databases. However, when it comes to NLIDB (natural language interface for databases), the entire system is often custom-made for a particular database. Overcoming the complexity and expressiveness of natural languages so that a single NLI can support a variety of databases is an unsolved problem. In this work, we show that it is possible to separate data specific components from latent semantic structures in expressing relational queries in a natural language. With the separation, transferring an NLI from one database to another becomes possible. We develop a neural network classifier to detect data specific components and an adversarial mechanism to locate them in a natural language question. We then introduce a general purpose transfer-learnable NLI that focuses on the latent semantic structure. We devise a deep sequence model that translates the latent semantic structure to an SQL query. Experiments show that our approach outperforms previous NLI methods on the WikiSQL [49] dataset, and the model we learned can be applied to other benchmark datasets without retraining.Keywords
This publication has 28 references indexed in Scilit:
- Compositional Semantic Parsing on Semi-Structured Tables2015
- Building a Semantic Parser Overnight2015
- Knowledge discovery in medicine: Current issue and future trendExpert Systems with Applications, 2014
- Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation2014
- Glove: Global Vectors for Word Representation2014
- Word-Based Dialog State Tracking with Recurrent Neural Networks2014
- The application of data mining techniques in financial fraud detection: A classification framework and an academic review of literatureDecision Support Systems, 2011
- The Hidden Information State model: A practical framework for POMDP-based spoken dialogue managementComputer Speech & Language, 2010
- Application of data mining techniques in customer relationship management: A literature review and classificationExpert Systems with Applications, 2009
- Natural language interfaces to databases – an introductionNatural Language Engineering, 1995