Impact of intellisense on the accuracy of Natural Language Interface to Database

Natural Language Interface to Database (NLIDB) is an interface which allows a naive user to ask query to database in his own language. The interface then translates the natural language query into SQL query and retrieves data from the database. NLIDB is not a recently rising topic. Many different approaches have been proposed in past for the development of an NLIDB but none could be used commercially. The main reasons are that the highest accuracy achieved is 96% and not all the concepts of SQL are mapped to them. Even a 4% error can cause massive damage to a business. Thus this paper proposes the design of GUI, implementation of intellisense and an error handling module for Natural Language Interface to Database to achieve 100% accuracy. We have enhanced our intellisense through machine learning so that it could generate suggestions based on previously typed words to help frame a complete and correct query. Traditional intellisenses show list of words whose initial letters are typed by the user and thus help in typing correct words but not correct query.