Rethinking the Detection of Child Sexual Abuse Imagery on the Internet
Open accessThe DOI matches its Crossref record; 9 of the 9 references it cites were found in a registry or an index.
Abstract
Over the last decade, the illegal distribution of child sexual abuse imagery (CSAI) has transformed alongside the rise of online sharing platforms. In this paper, we present the first longitudinal measurement study of CSAI distribution online and the threat it poses to society's ability to combat child sexual abuse. Our results illustrate that CSAI has grown exponentially-to nearly 1 million detected events per month-exceeding the capabilities of independent clearinghouses and law enforcement to take action. In order to scale CSAI protections moving forward, we discuss techniques for automating detection and response by using recent advancements in machine learning.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
FaceNet: A unified embedding for face recognition and clustering
Schroff, F., Kalenichenko, D., & Philbin, J. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298682 · Cited by 15,267
Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
Szegedy, C., Ioffe, S., Vanhoucke, V., & Alemi, A. (2017) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v31i1.11231 · Cited by 4,636 · cited 1×
“• Scene information: Image classifiers can automatically report objects that are present in a scene [39, 40], an image’s environment [35, 45], and potentially identify landmarks that will help locate the region where an abusive image originated [43].”
BackgroundImpact of sexual abuse on children: A review and synthesis of recent empirical studies
Kendall-Tackett, K. A., Williams, L. M., & Finkelhor, D. (1993) · Psychological Bulletin · doi:10.1037/0033-2909.113.1.164 · Cited by 2,537 · cited 1×
“Even after escaping abusers, victims face immediate and lasting consequences such as a heightened risk of depression, substance abuse, and suicide [5, 23].”
BackgroundA Global Perspective on Child Sexual Abuse: Meta-Analysis of Prevalence Around the World
Stoltenborgh, M., van IJzendoorn, M. H., Euser, E. M., & Bakermans-Kranenburg, M. J. (2011) · Child Maltreatment · doi:10.1177/1077559511403920 · Cited by 2,010
Ten-Year Research Update Review: Child Sexual Abuse
PUTNAM, F. W. (2003) · Journal of the American Academy of Child & Adolescent Psychiatry · doi:10.1097/00004583-200303000-00006 · Cited by 1,488 · cited 1×
“These children endure in-decent exposure, forced intercourse, and sex trafficking [33].”
A convolutional neural network cascade for face detection
Li, H., Lin, Z., Shen, X., Brandt, J., & Hua, G. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7299170 · Cited by 1,390
The prevalence of child sexual abuse in community and student samples: a meta-analysis
Pereda, N., Guilera, G., Forns, M., & Gómez-Benito, J. (2009) · Clinical Psychology Review · doi:10.1016/j.cpr.2009.02.007 · Cited by 1,068 · cited 1×
“Child sexual abuse is a horrific crime affecting an estimated 9–19.7% of girls and 3–7.9% of boys [2, 31, 38].”
InfluentialThe current prevalence of child sexual abuse worldwide: a systematic review and meta-analysis
Barth, J., Bermetz, L., Heim, E., Trelle, S., & Tonia, T. (2013) · International Journal of Public Health · doi:10.1007/s00038-012-0426-1 · Cited by 1,063
Quantifying paedophile queries in a large p2p system
Latapy, M., Magnien, C., & Fournier, R. (2011) · 2011 Proceedings IEEE INFOCOM · doi:10.1109/infcom.2011.5935191 · Cited by 20 · cited 1×
“In terms of detection and disruption, research has focused either on flagging queries for CSAI content based on known keywords [24, 36] or on targeting major seeders to prevent further distribution [17].”
Background
Cited by
An Overview of Perceptual Hashing
Farid, H. (2021) · Journal of Online Trust and Safety · doi:10.54501/jots.v1i1.24 · Cited by 70
“Within a fully E2EE system, content passing through the provider remains encrypted and inaccessible to PhotoDNA and the like.”
MethodInfluentialKnack, N., Holmes, D., & Fedoroff, J. P. (2020) · Behavioral Sciences & the Law · doi:10.1002/bsl.2450 · Cited by 35
Steel, C. M. S., Newman, E., O’Rourke, S., & Quayle, E. (2020) · Forensic Science International: Digital Investigation · doi:10.1016/j.fsidi.2020.300971 · Cited by 33
Bleakley, P., Martellozzo, E., Spence, R., & DeMarco, J. (2024) · European Journal of Criminology · doi:10.1177/14773708231181361 · Cited by 16
Metadata
- Authors
- Bursztein, E., Clarke, E., DeLaune, M., Elifff, D. M., Hsu, N., Olson, L., Shehan, J., Thakur, M., Thomas, K., & Bright, T.
- Venue
- The World Wide Web Conference
- Publisher
- ACM
- Published
- 2019-05-13
- Pages
- 2601-2607
- OpenAlex
- W2902726641
- Cited by
- 80
- References
- 9
- FWCI
- 4.01
- Type
- conference paper
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.