Record Deduplication By Evolutionary Means

Created by W.Langdon from gp-bibliography.bib Revision:1.7615

@Misc{oai:CiteSeerX.psu:10.1.1.604.610,

• author = "Marco Modesto and Moises G. {de Carvalho} and Walter {dos Santos}",
• title = "Record Deduplication By Evolutionary Means",
• howpublished = "CiteSeerX",
• year = "2002?",
• address = "Departamento de Ciencia da Computacao, Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil",
• keywords = "genetic algorithms, genetic programming",
• annote = "The Pennsylvania State University CiteSeerX Archives",
• bibsource = "OAI-PMH server at citeseerx.ist.psu.edu",
• language = "en",
• oai = "oai:CiteSeerX.psu:10.1.1.604.610",
• rights = "Metadata may be used without restrictions as long as the oai identifier remains attached to it.",
• URL = "http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.604.610",
• URL = "http://homepages.dcc.ufmg.br/~nivio/cursos/pa06/seminarios/seminario16/seminario16.pdf",
• size = "5 pages",
• abstract = "Identifying record replicas in digital data repositories is a key step to improve the quality of content and services available, as well as to yield eventual sharing efforts. Several deduplication strategies are available, but most of them rely on manually chosen settings to combine evidence used to identify records as being replicas. In this work, we present the results of experiments we have carried out with a Machine Learning approach for the deduplication problem. Our approach is based on Genetic Programming (GP), that is able to automatically generate similarity functions to identify record replicas in a given repository. The generated similarity functions properly combine and weight the best evidence available among the record fields in order to tell when two distinct records represent the same real-world entity. On a previous work, fixed similarity functions were associated to each evidence. On the present work, the GP will be also used to choose the best evidence and similarity functions associations. The results of the experiments show that our approach outperforms the baseline method by Fellegi and Sunter. It also outperformed the previous GP results, using fixed evidence associations when identifying replicas in a data set containing researcher's personal data.",
• notes = "Oct 2017 reference by http://homepages.dcc.ufmg.br/~nivio/cursos/pa06/seminarios/",

}

Genetic Programming entries for Marco Modesto Moises G de Carvalho Walter dos Santos

Citations