Research map / Adversarial robustness and security / Attacks on text and graphs
One-Class Adversarial Nets for Fraud Detection
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use complementary samples rather than regular-GAN samples to train the discriminator (ablation) | OCAN · Wikipedia vandal detection | F1: 0.8701 ± 0.0267 -> 0.9010 ± 0.0228 | Helped |
| use complementary samples rather than regular-GAN samples to train the discriminator (ablation) | OCAN · Early Wikipedia vandal detection | F1: 0.7874 -> 0.8459 | Helped |
| use LSTM-Autoencoder user representations instead of raw features | one-class classifiers on Wikipedia vandal detection · Wikipedia vandal detection | F1: 0.6845 ± 0.0184 -> 0.8150 ± 0.0163; 0.7023 ± 0.0193 -> 0.8326 ± 0.0158; 0.7919 ± 0.0040 -> 0.7789 ± 0.0064 | Mixed |
| train M-LSTM with fewer labeled vandals | M-LSTM · Early Wikipedia vandal detection | F1: 0.7951 -> 0.0000 | Hurt |
| use transaction representations instead of raw transaction features | OCAN · Credit card fraud detection | F1: 0.8416 ± 0.0330 -> 0.8656 ± 0.0220 | Helped |
| replace the LSTM-Autoencoder with a regular autoencoder for transaction data | OCAN · Credit card fraud detection | F1 | No clear effect |
Closest papers by meaning
- Toward More Generalized Malicious URL Detection Models (AAAI 2024 · similarity 0.85)
- Unveiling the Threat of Fraud Gangs to Graph Neural Networks: Multi-Target Graph Injection Attacks Against GNN-Based Fraud Detectors (AAAI 2025 · similarity 0.84)
- Defense Against Model Stealing Based on Account-Aware Distribution Discrepancy (AAAI 2025 · similarity 0.82)
- DiG-In-GNN: Discriminative Feature Guided GNN-Based Fraud Detector against Inconsistencies in Multi-Relation Fraud Graph (AAAI 2024 · similarity 0.82)
- Unlearn and Burn: Adversarial Machine Unlearning Requests Destroy Model Accuracy (ICLR 2025 · similarity 0.82)
- Pre-trained Behavioral Model for Malicious User Prediction on Social Platform (AAAI 2025 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.