Research on the Technology of Generating Single-Table sql Query Sentences in Chinese Natural Language

With the rapid development of the Internet nowadays, the amount of data has become larger and larger. It is easier to save data in the database, but finding the information people want from the database has become an urgent problem. The main task of the natural language generating SQL query statement task is to convert the natural language into an executable SQL query statement. However, how to convert natural language into SQL queries has become a difficult task for this task. This research mainly uses the template-filled SQLNet model as the benchmark model to design a model from Chinese natural language to SQL query statements. At the same time, two modules, select num and where relation, are added to the Select clause and the where clause, respectively. At the same time, modify the condition value module in the where clause and then predict the condition value after adding the column type information. This Chinese NL2SQL deep learning model was experimentally verified on an open Chinese data set TableQA. The experimental results show that the model can effectively convert Chinese natural language into SQL query sentences.