Abstract

Generating structural query language (SQL) queries from natural language is along-standing open problem. Answering a natural language question about adatabase table requires modeling complex interactions between the columns ofthe table and the question. In this paper, we apply the synthesizing approachto solve this problem. Based on the structure of SQL queries, we break down themodel to three sub-modules and design specific deep neural networks for each ofthem. We employ the bidirectional attention mechanisms and character- levelembedding with convolutional neural networks (CNNs) to improve the result.Experimental evaluations show that our model achieves the state-of-the-artresults in WikiSQL dataset.

 

Quick Read (beta)