How would you detect fraud rings (colluding accounts)?
hardAnswer
- (1) Build interaction graph: accounts as nodes, shared devices/IPs/transactions as edges.
- (2) Community detection (Louvain, Leiden, HDBSCAN on node embeddings).
- (3) Node2Vec / GraphSAGE embeddings → cluster.
- (4) Anomalous dense subgraphs / motifs.
- (5) Combine with per-account anomaly scores.
- Standard in banking / crypto / marketplace fraud (Uber, Airbnb).
- GNN-based methods (GraphSAGE + supervised head where labels exist) common.
Check yourself — multiple choice
- Same as per-account
- Build interaction graph + community detection (Louvain/Leiden) or GNN embed + cluster + anomalous dense subgraphs; standard for banking / marketplace fraud rings
- Random
- Not real
Fraud rings: graph + community detection / GNN embed + anomalous subgraphs.
#applications#anomaly-detection
Practise Unsupervised Learning
214 interview questions in this topic.
Related questions
- Anomaly detection in time series — what changes?
- How is drift detection an unsupervised problem?
- Covariate drift vs concept drift vs label drift — the differences.
- Scan statistics — when do you use them?
- CUSUM change-point detection — how does it work?
- How do you handle 99.9% normal / 0.1% anomaly training data?