Semi-Supervised Supply Chain Fraud Detection with Unsupervised Pre-Filtering
Fatemeh Moradi, Mehran Tarif, Mohammadhossein Homaei · 2025
Read Semi-Supervised Supply Chain Fraud Detection with Unsupervised Pre-Filtering by Fatemeh Moradi, Mehran Tarif, Mohammadhossein Homaei (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-06574.txt.
Description
Fraud hides well inside modern supply chains: networks are vast, and confirmed fraud labels are rare, so detectors trained only on labels start weak and drown in false alarms. Fatemeh Moradi, Mehran Tarif, and Mohammadhossein Homaei propose a two-phase detector in 2025. First, an unsupervised method, Isolation Forest, flags odd records and shrinks the pile that needs closer review.
A support vector machine then trains on the labeled cases plus its own most confident guesses, a self-training loop that grows the labeled set safely. On the DataCo smart supply chain dataset, the system reached an F1 score of 0.817 while keeping false positives under three percent. The authors say the pairing of unsupervised pre-filtering with semi-supervised refinement suits real-world limits, while noting that shifting fraud patterns and comparisons with deep learning remain open work.
Catalog record
| CREATOR | Fatemeh Moradi, Mehran Tarif, Mohammadhossein Homaei |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 26 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-06574