| Summary: | There are different methodologies for identification of fuzzy duplicate records in the process of data cleaning for data warehouse and data mining. The methodologies for duplicate record identification can be classified into three groups: blocking methods, windowing methods, and semantic methods. The article specifically focuses on semantic methods and describes Semantic-Syntactic Method for fuzzy duplicate record identification. Based on the conducted testing, comparative analysis is presented of the results obtained through the Semantic-Syntactic Method and two other standard methods over a selected data set. In the end, the article presents conclusions with regard to the quality and efficiency of the Semantic-Syntactic Method, as well as suggestions for future research in this field. Keywords: Duplicate records, Fuzzy, Semantic, Syntactic, Blocking, Windowing
|
|---|
