Authors:
Gabrielle Karine Canalle
;
Bernadette Farias Lóscio
and
Ana Carolina Salgado
Affiliation:
Federal University of Pernambuco, Brazil
Keyword(s):
Attribute Selection, Entity Resolution, Data Integration.
Related
Ontology
Subjects/Areas/Topics:
Coupling and Integrating Heterogeneous Data Sources
;
Databases and Information Systems Integration
;
Enterprise Information Systems
Abstract:
Data integration is an essential task for achieving a unified view of data stored in heterogeneous and distributed data sources. A key step in this process is the Entity Resolution, which consists of identifying instances that refer to the same real-world entity. In general, similarity functions are used to discover equivalent instances. The quality of the Entity Resolution result is directly affected by the set of attributes selected to be compared. However, such attribute selection can be challenging. In this context, this work proposes a strategy for selection of relevant attributes to be considered in the process of Entity Resolution, more precisely in the instance matching phase. This strategy considers characteristics from attributes, such as quantity of duplicated and null values, in order to identify the most relevant ones for the instance matching process. In our experiments, the proposed strategy achieved good results for the Entity Resolution process. Thus, the attributes
classified as relevant were the ones that contributed to find the greatest number of true matches with a few incorrect matches.
(More)