Natural Language Interface to Database by Regular Expression Generation
Mithila Kundu Barsha, K. M. Azharul Hasan, Irfat Ara · 2021
Many of the information systems are static because the codes of the Structured Query Language (SQL) are written by the experts, and adding new queries is impossible without the help of experts. Natural Language Interface to Database (NLID) solves this problem. NLID is an idea where non-experts can retrieve information from a database by simply writing queries using natural language text. In NLID, a natural language input is executed to generate an appropriate SQL statement, and this SQL statement is executed to retrieve the information from the underpinning database as output. Therefore, retrieval from underpinning data by NLID is an important research area in database and NLP for easy and convenient information retrieval. In this paper, we propose an NLID system for fast and convenient information retrieval using pattern matching. The pattern matching is performed using the dynamic generation of regular expressions. We develop a student information system for the case study and measure the performance of the proposed approach.