Comparison of record linkage methods

Record linkage is an important tool to enhance database integration. This even more valuable in a scenario with hefty budget cuts and growing drop response rate traditional surveys. strategy makes it possible expand the crossing alternatives variables not present original base. However, there are many different data pairing methods exposed literature. In this sense, objective of paper compare well-known record linkage. The comparison was made synthetic dataset. To methods, adopted quantitative approach based on Precision, Recall, F-Statistics metrics, using two functions: Levenshtein Jaro-Winkler. Among six types classifiers analyzed, supervised had best results.