Modelling the Convergence of Victims, Offenders and Media Files in Child Sexual Abuse Investigations

• In Event: Methods for Researching and Responding to Sexual Violence

Abstract

Child sexual abuse investigations involving the seizure of electronic media (videos, images, and audio recordings) can require the painstaking and labor-intensive forensic examination of thousands of files to locate information about victims and/or offenders. The often-graphic nature of the content can also have serious adverse consequences for investigator mental health and has accordingly led to the adoption of strategies that seek to limit exposure where possible. While the adoption of such strategies is necessary, they may also restrict the quantity and quality of intelligence data that can be extracted and analyzed from seized evidence. This paper introduces an automated software system designed by the research team to overcome this problem. In particular, our system proposes a data processing methodology to extract, and match multiple biometric modalities (face and voice), and automatically predict the occurrence of distinct subjects (i.e. victims or offenders) contained in, and the connections across media files. We demonstrate this capability using a collection of 10,561 CSAM files seized by Australian law enforcement, and show how match data can be analyzed using additional modelling approaches to rapidly identify subjects that may be ‘core’ or ‘peripheral’ to an investigation. Directions for future research are also discussed.

Authors

• Bryce Westlake, San Jose State University

• Russell Brewer, University of Adelaide

• Thomas Swearingen, University of Adelaide