Non-English Natural Language Interface to Databases: A Systematic Review

A natural language interface (NLI) facilitates communication between a user and a system using a natural language used by a human as a medium. Instead of a complex user interface with numerous controls, the user enters input via natural language in the form of text, speech, signs, etc. The system responds via natural language in the form of text, speech, signs, etc. The use of a natural language interface has been increasing. To interact with a dataset or a database, such a system or interface is not always available. To access data, users need the knowledge of Structured Query Language (SQL) or need to learn an alternative graphical tool or database browser with varying limitations. For non-English speakers, it may create further issues in accessing data. English users can access the database using the natural language interface to database (NLIDB) which converts natural language to English-like Structured Query Language (SQL). For non-English users, the query in non-English language needs to be converted to SQL directly or via the English language as an intermediary. Researchers have proposed NLIDB systems for native speakers of many languages. In this research, we present results from a systematic review of non-English NLIDB with their various approaches. This research aims to assist future NLIDB researchers in understanding the current state and to identify the research gaps.