Xml Document Clustering: An Algorithmic Perspective

The wide use of XML as the de facto standard of storing and exchanging information through Internet has led a wide spectrum of heterogeneous applications to adopt XML as their information representation model. The heterogeneity of XML data sources has brought up the problem of efficiently clustering a set of XML documents. However, traditional clustering algorithms cannot be applied due to the semistructured nature of XML, which contains both structure and content features. Hence, special techniques should be used that would take into account the XML semantics in order to address the problem of XML clustering. The described approaches, based on either the structure or the content or both, manage to successfully address the problem and can be applied efficiently in real-world applications.