Enhancing SQL Query Generation from Natural Language Inputs Using Deep Learning Models with NLI-RDB

Abstract

This paper proposes a new framework that is hoped will increase the effectiveness of deep learning models in translating natural language inputs into SQL queries. The proposed framework also effectively addresses the problem of capturing the semantic meaning of natural language queries through the combination of RNNs and transformer models to generate high-quality SQL translations. The combination of the advantages of RNNs for sequential data and transformers for context task allows the system to generate queries with a relatively high degree of accuracy. A comparison of the proposed framework with several popular models quantitatively shows the effectiveness of the proposed method in terms of both accuracy and performance, which makes it an excellent addition to the existing literature on NLP and databases. The obtained results show the usefulness of this hybrid deep learning model for the described database querying process and indicate that this model can be used for the development of application programs that will address the problem of manual query formulation and facilitate user interaction with database systems through natural language interfaces.