Abstract
In information technology, one predominant requirement is to design interfaces in more natural way in order to enable users to interact with computers in an easy way. Natural Language Interfaces to Databases (NLIDBs) is one of the mechanisms to achieve this goal. This paper is based on the previous work on NLIDBs for Urdu/English language in which an attribute value (AV) Mapping Algorithm was introduced. This algorithm uses semantic dictionary to efficiently map natural language queries to SQL queries and minimize its transformation time, but suffers from the ambiguity problem occurring in the queries. To overcome this problem, this study focused on the identification and resolution of ambiguous queries. New enhancement in the AV Mapping algorithm was made to improve the processing and enabling the algorithm to identify and disambiguate many ambiguous cases. All the techniques suggested in the AV Mapping Algorithm are applicable to any other language that has subject-verb-object pattern. That is, the algorithm is based on the identification and treatment of semantic tokens using semantic dictionary. The algorithm is implemented in Visual C#.NET and tested on Student Information System and Employee Information System databases. The accuracy of correct mapping is 85%
