Anybody interested in entity resolution--the process of determining whether two references to an entity refer to the same object or different objects--should read this valuable paper. The authors assume that entity resolution and information quality converge, and that this convergence should be studied in graduate-level courses.
The authors state that solving information quality problems is a prerequisite for properly preparing the reference sources for entity resolution decisions. This approach contributes to the quality of an information product created from several information sources. Further, they describe a core course subject dealing with the entity identity, information quality, and record linkage models and techniques. Readers are also presented with the authors’ notions on practical exercises that focus on the course material and the use of various tools, such as the OYSTER (Open sYSTem Entity Resolution) system, an open-source software development project that can be used as a teaching tool to give students experience performing entity resolution.
The concluding remarks state that students benefit from a series of exercises that help them understand how improvements in the quality of information sources can improve overall results in the entity resolution process.
I found the paper interesting and recommend it to students and professionals working in any sector or field where information integration is a core issue.