Errors of identifiers in anonymous databases: impact on data quality

Pombinho, Paulo; Cavique, Luís; Correia, Luís

http://hdl.handle.net/10400.2/12904

Utilize este identificador para referenciar este registo.

Nome:	Descrição:	Tamanho:	Formato:
paper SOCO.pdf		409.78 KB	Adobe PDF	Ver/Abrir

Contacte-nos

Autores

Pombinho, Paulo

Cavique, Luís

Correia, Luís

Resumo(s)

Data quality is essential for a correct understanding of the concepts they represent. Data mining is especially relevant when data with inferior quality is used in algorithms that depend on correct data to create accurate models and predictions. In this work, we introduce the issue of errors of identifiers in an anonymous database. The work proposes a quality evaluation approach that considers individual attributes and a contextual analysis that allows additional quality evaluations. The proposed quality analysis model is a robust means of minimizing anonymization costs.