CoXML: Cooperative XML Query Answering?

As the World Wide Web becomes a major means in disseminating and sharing information, there has been an exponential increase in the amount of data in web-compliant format such as HyperText Markup Language (HTML) and Extensible Markup Language (XML). XML is essentially a textual representation of the hierarchical (tree-like) data where a meaningful piece of data is bounded by matching starting and ending tags, such as and . Due to the simplicity of XML as compared to SGML and the expressiveness of XML as compared to HTML, XML has become the most popular format for information representation and data exchange. To cope with the tree-like structure in the XML model, many XML-specific query languages have been proposed(e.g., XPath 1 and XQuery [4]). All these query languages aim at the exact matching of query conditions. Answers are found when those XML documents match the given query condition exactly. However, this may not always be the case in the XML model. To remedy this condition, we propose a cooperative query answering framework that derives approximate answers by relaxing query conditions to less restricted forms. Query relaxation has been successfully used in relational databases (e.g., [7], [6], [5], [12], [24]) and is important for the XML model because: