Identification and Localization of Data Types within Large-Scale File Systems

Abstract

This research examines the application of statistical analysis techniques for the identification of data types embedded within a file to assist analysts with the location of data, potentially relevant to criminal activity. The results show that the statistical analysis can effectively aid identification of the types of data embedded in a file and the approximate location of these data types. This analysis identifies component data types, irrespective of the type of file being analyzed. When applied, this technique will allow analysts to more effectively and efficiently locate relevant data on a hard drive, especially on today's particularly large hard drives