Cyber child pornography is an increasingly visible problem in society today. With the growth in home Personal Computer (PC) usage and more readily available access to the World Wide Web over the past decade, child pornographers have found a convenient venue for sharing horrific pictures of children being sexually abused. Also, police and lawyers around the globe have found that detecting and prosecuting cyber child pornographers have become onerous chores, often with a high failure rate of placing perpetrators behind bars. The methods currently employed by law enforcement officers to combat cyber child pornography may be considered to be primitive and inefficient. In this paper, we review the major social, legal, and technological issues facing citizens, lawmakers, and the police regarding cyber child pornography. We also propose a new technological approach for combating online child pornography. In particular, we propose a source address reputation system based on Bloom filters and a novel classification system utilizing a stochastic weak estimator, coupled with a linear classifier. We believe that our proposed method for identifying offensive online material would be attractive to law enforcement globally, because it can be implemented with acceptable overheads.
Financed by the National Centre for Research and Development under grant No. SP/I/1/77065/10 by the strategic scientific research and experimental development program:
SYNAT - “Interdisciplinary System for Interactive Scientific and Scientific-Technical Information”.