A New Architecture for NLIDB Using Local Appropriator Engine for SQL Generation

• Published in: International journal of advanced research in computer science 2011-11, Vol.2 (6)
• Main Authors: Enikuomehin, Oluwatoyin Ayokunle, S, Sadiku J
• Format: Article
• Language: English
• Subjects: Algorithms; Computer science; Keywords; Language; Natural language; Queries; Relational data bases; Semantics
• ISSN: 0976-5697

In this paper, we present the design and implementation a special type of Natural Language Interface using the concept of Local Appropriator (LA). The purpose of natural language introduction to database is to allow user post questions coined "natural language" and get discriminated answers which form the result of their query in terms of tables. A Local Appropriator Natural Language Interface, the LANLI, parses semantically, a natural language on a relational database by building the corresponding SQL. The LANLP is a new generation of NL2DB which uses a Syntactic-Semantic (SySe) tree to generate a matching algorithm for its transformation using word net for semantic analysis and a parse tree algorithm for its syntactic analysis. The generated tree is fuzzified to handle ambiguity contained in word. For expert users, this advancement could be likened to the development of an automated structure query language (sql) generator. The essence of this paper is to present a new architecture for querying databases that provide users with the capabilities of extracting information contained in the relational database. Users do not have to learn the process of formulating a query through sql.