Method and Software for Solving the Problem of Fuzzy Matching of Records in Relative Databases

Abstract

The fuzzy search computing technology implemented for MSSQL database with using .NET plattform is considered in the work. The proposed technology is based on the calculation of the Levenstein distance. It is shown, that it is advisable to use the technology to search for names in an existing dictionary, where there is a high probability that the user may make a typo or make a mistake on at least one character. To confirm the effectiveness of the proposed concept, a demo program was created with the adding functions to provide a fuzzy comparison of attributes by data type. This approach is a more effective solution of the problem through the use of functions for the primary conversion of data according to the selected format, and after the converted data is compare using the selected method for fuzzy matching of attributes in a relational database.