Using information quality for the identification of relevant web data sources

Abstract

In the last decade, applications that make use of data sources available on the Web have experienced a huge growth. One of the main problems regarding that consists in finding the most relevant data sources for a given application. In a general way, a data source is considered relevant when it contributes for answering queries submitted to the application. However, it may happen that a specific data source contributes for answering an application query but the answer provided by the data source does not really meet the user requirements. This may occur because the data source has generic data and the user wants more specific data, for example. On the other hand, some data sources may have data of poor quality, i.e., the data may be outdated, incomplete or incorrect. In such cases, it is not enough just to find data sources that can answer to the application queries. It is also important to check if the available data also meet the user needs. In this paper, we discuss such problem and we propose an approach, based on Information Quality (IQ), to help the evaluation of the relevance of a Web data source for domain-specific applications. We also present an example illustrating how our proposal can be used to enhance this evaluation.