Thesis
Generative AI for Fraud Detection in PropTech
Can adversarially trained generative models detect fraudulent property transactions that rule based and supervised systems miss?
- Kind
- Thesis
- Year
- 2023
- Status
- Complete
01
Abstract
An investigation into generative adversarial approaches for anomaly detection in real estate transaction monitoring. Property fraud is rare, adversarial, and constantly changing shape, which is precisely the setting where supervised classifiers trained on historical labels perform worst. The work trains a model of legitimate transaction structure and treats reconstruction difficulty as the fraud signal, then examines where that assumption holds and where an adversary can exploit it.
02
Methods
- Generative adversarial networks
- Anomaly detection
- PyTorch
- Imbalanced learning