Temporal Constraints in Online Dating Fraud Classification

Harrison Bullock, Matthew Edwards

2023

Abstract

A number of automated systems attempt to combat online fraud through the application of classifiers created using machine learning techniques. However, online fraud is a moving target, and cybercriminals alter their strategies over time, causing a gradual decay in the effectiveness of classifiers designed to detect them. In this paper, we demonstrate the existence of this concept drift in an online dating fraud classification problem. Working with a dataset of real and fraudulent dating site profiles spread over 6 years, we measure the extent to which dating fraud classification performance may be expected to decay, finding substantial decay in classifier F1 over time, amounting to a decrease of more than 0.2 F1 by the end of our evaluation period. We also evaluate strategies for keeping fraud classification performance robust over time, suggesting mitigations that may be deployed in practice.

Paper Citation

in Harvard Style

Bullock H. and Edwards M. (2023). Temporal Constraints in Online Dating Fraud Classification. In Proceedings of the 9th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP, ISBN 978-989-758-624-8, pages 535-542. DOI: 10.5220/0011689000003405

in Bibtex Style

@conference{icissp23,
author={Harrison Bullock and Matthew Edwards},
title={Temporal Constraints in Online Dating Fraud Classification},
booktitle={Proceedings of the 9th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP,},
year={2023},
pages={535-542},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011689000003405},
isbn={978-989-758-624-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 9th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP,
TI - Temporal Constraints in Online Dating Fraud Classification
SN - 978-989-758-624-8
AU - Bullock H.
AU - Edwards M.
PY - 2023
SP - 535
EP - 542
DO - 10.5220/0011689000003405