Resident data pattern analysis using sector clustering for storage drive forensics

Abstract

Storage drives are huge reservoirs of digital evidence. The acquisition and examination of storage drives for evidentiary artifacts require enormous amounts of manual effort and computing resources, leading to huge case backlogs. This chapter describes a forensic triage methodology that leverages random sampling and unsupervised clustering to provide insights about the regions of interest on a storage drive. The number of sector samples to be evaluated during triage for legitimate inferences to be drawn about drive content is also discussed. Experiments involving storage drives of various capacities illustrate the effectiveness and utility of the extracted patterns for rapid drive triage.