Abstract
Information marked up as XML data is becoming increasingly pervasive as a part of business-to-business electronic transactions. A possible threat to the continued growth of XML in this domain is that data mining technology may be applied to XML documents in order to reveal sensitive knowledge. This paper presents a methodology for hiding sensitive knowledge in XML documents in the context of association mining algorithms. This methodology involves identifying the sensitive knowledge within the document, formulating an appropriate set of security policies, and finally sanitizing the document to hide the sensitive knowledge.