Crime detection is an area of vital importance in police department. Crime rate are rapidly changing and improved analysis enables discerning hidden pattern of crime, if any, without any explicit prior knowledge of these pattern. Detecting crime from data analysis can be difficult because daily activities of criminal generate large amounts of data. The police records exist in various formats and the quality of analysis greatly depends on the background knowledge of the analyst. This paper proposes a simple correlation clustering algorithm which aims at finding illegal activities of professional identity fraudsters based on knowledge discovered from their own histories.