SAGE: Scalable Automatic Gating Ensemble for Confident Negative Harvesting in Fraud Detection
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911698732974080 |
|---|---|
| author | Tubati, Sudheer Goyal, Amit |
| author_facet | Tubati, Sudheer Goyal, Amit |
| contents | Music streaming fraud, where bad actors artificially inflate stream counts to manipulate chart rankings and royalty payments, poses a significant threat to streaming services and legitimate content creators. Traditional fraud detection approaches struggle with a critical challenge: many legitimate edge cases, including super-fans and sleep-music sessions, exhibit activity patterns that closely mimic those of coordinated fraud. We present SAGE, a novel counterfactual-aware negative harvesting approach that combines SimHash-based stratified sampling with a modular gating ensemble for confident negative identification from unlabeled data. Our ensemble architecture employs pluggable statistical gates (currently instantiated with Mahalanobis distance and k-NN density) with configurable voting thresholds enabling adaptive precision-recall trade-offs. This addresses the representation bias problem in Positive-Unlabeled learning by ensuring comprehensive coverage of rare behavioral cohorts through floor-constrained sampling. Evaluation demonstrates strong precision and recall on held-out data. The approach generalizes across fraud detection domains, achieving strong performance on both customer-level and artist-level fraud without modification to the core methodology. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_20157 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SAGE: Scalable Automatic Gating Ensemble for Confident Negative Harvesting in Fraud Detection Tubati, Sudheer Goyal, Amit Machine Learning Cryptography and Security Information Retrieval 68T05 (Primary), 62H30, 68T10 (Secondary) Music streaming fraud, where bad actors artificially inflate stream counts to manipulate chart rankings and royalty payments, poses a significant threat to streaming services and legitimate content creators. Traditional fraud detection approaches struggle with a critical challenge: many legitimate edge cases, including super-fans and sleep-music sessions, exhibit activity patterns that closely mimic those of coordinated fraud. We present SAGE, a novel counterfactual-aware negative harvesting approach that combines SimHash-based stratified sampling with a modular gating ensemble for confident negative identification from unlabeled data. Our ensemble architecture employs pluggable statistical gates (currently instantiated with Mahalanobis distance and k-NN density) with configurable voting thresholds enabling adaptive precision-recall trade-offs. This addresses the representation bias problem in Positive-Unlabeled learning by ensuring comprehensive coverage of rare behavioral cohorts through floor-constrained sampling. Evaluation demonstrates strong precision and recall on held-out data. The approach generalizes across fraud detection domains, achieving strong performance on both customer-level and artist-level fraud without modification to the core methodology. |
| title | SAGE: Scalable Automatic Gating Ensemble for Confident Negative Harvesting in Fraud Detection |
| topic | Machine Learning Cryptography and Security Information Retrieval 68T05 (Primary), 62H30, 68T10 (Secondary) |
| url | https://arxiv.org/abs/2605.20157 |