With the volume of digital forensic evidence rapidly increasing, this paper proposes a data reduction and data mining framework that incorporates a process of reducing data volume by focusing on a subset of information.
The volume of digital forensic evidence is rapidly increasing, leading to large backlogs. In this paper, a Digital Forensic Data Reduction and Data Mining Framework is proposed. Initial research with sample data from South Australia Police Electronic Crime Section and Digital Corpora Forensic Images using the proposed framework resulted in significant reduction in the storage requirements—the reduced subset is only 0.196 percent and 0.75 percent respectively of the original data volume. The framework outlined is not suggested to replace full analysis, but serves to provide a rapid triage, collection, intelligence analysis, review and storage methodology to support the various stages of digital forensic examinations. Agencies that can undertake rapid assessment of seized data can more effectively target specific criminal matters. The framework may also provide a greater potential intelligence gain from analysis of current and historical data in a timely manner, and the ability to undertake research of trends over time.