1 Introduction
Knowledge acquisition and management are key to developing successful applications on the semantic Web [1] aiming at enabling people and computers to interact more naturally. While some structured and semistructured data and information (i.e., data in relational database) can be transformed into structured knowledge (i.e., RDF
http://www.w3.org/RDF/.
data sets) for semantic Web applications, the transformation of other data might be difficult. This is due to the shortage of general methods for generating structured or semistructured data. Also, many data sources choose to protect key data from being accessed by the public, so as to maintain their competitive advantages.