Automatically Dismantling Online Dating Fraud
- Submitting institution
-
King's College London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 111096269
- Type
- D - Journal article
- DOI
-
10.1109/TIFS.2019.2930479
- Title of journal
- IEEE Transactions on Information Forensics and Security
- Article number
- 8768406
- First page
- 1128
- Volume
- 15
- Issue
- -
- ISSN
- 1556-6013
- Open access status
- Compliant
- Month of publication
- July
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
5
- Research group(s)
-
-
- Citation count
- 4
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Online dating fraud has been a long-standing problem, with annual losses reaching $143M (https://www.ftc.gov/news-events/press-releases/2019/02/new-ftc-data-spotlight-details-big-jump-losses-complaintsabout). Scammers craft appealing fake profiles to effectively attract victims while evading spam-like detectors. The paper presents the first system tailored to automatically and accurately detect this fraud. The system outperforms humans by combining deep learning of features that capture idealised romantic beliefs from images and natural language. Robust to the omission of profile details, it performs at high accuracy (97%) with hold-out validation on a real-word dataset. The software is open-sourced (https://github.com/gsuareztangil/automatic-romancescam-digger), state-of-the-art, and led to invited talks including at NCSC and RICS stakeholder meetings (https://warwick.ac.uk/fac/sci/wmg/research/digital/csc/about/dapm/stakeholderevent/dapm_stakeholder_meeting_2018-11-22.pdf).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -