represent the adopted information quality evaluation criteria and factors able to represent users’ quality expectations (Pipino et al., 2002).
Published in Chapter:
A Web Metadata Based-Model for Information Quality Prediction
Ricardo Barros (COPPE - Federal University of Rio de Janeiro, Brazil), Geraldo Xexéo (COPPE - Federal University of Rio de Janeiro, Brazil), Wallace A. Pinheiro (COPPE - Federal University of Rio de Janeiro, Brazil), and Jano de Souza (COPPE - Federal University of Rio de Janeiro, Brazil)
Copyright: © 2008
|Pages: 20
DOI: 10.4018/978-1-59904-847-5.ch019
Abstract
Currently, in the Web environment, users have to deal with an enormous amount of information. In a Web search, they often receive useless, replicated, outdated, or false data, which, at first, they have no means to assess. Web search engines provide good examples of these problems: As reply from these mechanisms, users usually find links to replicated or conflicting information. Further, in these cases, information is spread out among heterogeneous and unrelated data sources, that normally present different information quality approaches. This chapter addresses those issues by proposing a Web Metadata-Based Model to evaluate and recommend Web pages based on their information quality, as predicted by their metadata. We adopt a fuzzy theory approach to obtain the values of quality dimensions from metadata values and to evaluate the quality of information, taking advantage of fuzzy logic’s ability to capture humans’ imprecise knowledge and deal with different concepts.