Plagiarisms are frequently occurring in the Computer Science courses, especially in computer programming. In this thesis, XML plagiarism detection model is introduced and XML is used as an intermediate data exchange mechanism in the suggested model. Since a procedural programming language is defined by its specific rules and it is well-structured form, we can generate an XML document from a program source based on the XML schema. As long as we can generate XML documents from given program sources, XML queries to extract information how they are similar to each other can be queried over the documents. This idea makes us encourage to moving our attention to plagiarism detection models. As the result of this study, XML Plagiarism Detection System (XPDec) has been developed. The plagiarism detection systems for programming can be classified into two main groups: attribute-counting based systems and structural metrics based systems. XPDec system uses the combined detection mechanism. In addition to the mechanism, XPDec system is adopting the XML query language suggested in XML plagiarism detection model to extract control sequences from XML documents. This mechanism increases the accuracy of the results. At the end of this thesis we shows that the XPDec system gives high accuracy results to finding similarities among the given source programs.