Finding Plagiarism by Evaluating Document Similarities
Jan Kasprzak, Michal Brandejs, Miroslav Křipač · 2009
In this paper we discuss the approach we have used for finding plagiarized passages of text during the PAN'09 plagiarism detection competition. We describe the existing anti-plagiarism system we use in the Czech National Archive of Graduate Theses. We then discuss the modifications to this system which have been necessary in order to fit the results to the competition rules. We also present a performance data of the described system, and the possible improvement for our production systems, which result from the code written for the PAN'09 competition.