An Approach for Developing Natural Language Interface to Databases Using Data Synonyms Tree and Syntax State Table

The basic idea addressed in this research is developing a generic, dynamic, and domain independent natural language interface to databases. The approach consists of two phases; configuration phase and operation phase. The former builds data synonyms tree based on the database being implemented. The idea behind this tree is matching the natural language words with database elements. The tree hierarchy contains the database tables, attributes, attribute descriptions, and all possible synonyms for each description. The latter phase contains a technique that implements syntax state table to extract the SQL components from the natural language user request. As a result the corresponding SQL statement is generated without interference of human experts.