SpeakQL Natural Language to SQL

Abstract

Incorporating SQL questions from normal language is a long-standing open issue and has been drawing in extensive intrigue as of late. Natural Language Interface (NLI) is the confluence of Natural Language Processing (NLP) and Human-Computer Interaction, which allows interaction between humans and computers through the utilization of Natural Language. Here we are gonna deal with the problem of automatic generation of Structured Query Language (SQL) queries. SQL is a database language for querying and manipulating relational databases. Despite the spectacular rise in the acceptance of relational databases, there is a fundamental limitaion to the ability to fetch data from those databases. One of the major reasons for this is the fact that the users of these relational databases need to comprehend convoluted structured query languages. In this body of work, we present an interface that allows users to interact with the databases using Natural Lanaguage as opposted to the conventional structure query languages.

Machine Learning System for Natural Language to SQL Translation

“…Additionally, the system described, similarly to Yu et al [7], does not follow the entire SQL vocabulary; keywords such as LIKE and OVER are absent. Shah et al [9] presented complies with the architectural progress from the seq2seq to the BERT model. Despite having an interesting elaborative idea, the results yield similar results to RAT-SQL's.…”

“…The main compression is, in this case, the ready-to-go formality of the application in which, upon entering the desired prompt, the query is executed immediately. In other words, Shah et al [9] outlines a very similar to RAT-SQL in its final attempts, which is coloured furthermore by an adequate interface for such a task.…”