Autoencoders for Financial Anomaly Detection: A Review of Architectures, Evaluation Practice and Deployment Gaps

Financial institutions must decide, within the milliseconds of a payment authorisation window and on evidence that is almost entirely unlabelled, whether a transaction is fraudulent: positives account for less than 0.2% of records, confirmatory labels arrive months later as chargebacks, and the gene...

Descripción completa

Guardado en:
Autores principales: Alexander Haro-Sarango, Diana Castillo‐Martínez, Silvia Cachay-Salcedo, Jessica Karina Saavedra-Vasconez, Rosa Salcedo-Dávalos, Briseida Sotelo-Castro, Carlos Saavedra-Vasconez
Formato: Reference entry
Lenguaje:inglés
Publicado: 2026
Materias:
autoencoder
financial anomaly detection
fraud detection
one-class classification
reconstruction error
card payments
transaction monitoring
anti-money laundering
insurance claims
audit logs
generative models
resampling
evaluation metrics
temporal drift
class imbalance
Anomaly Detection Techniques and Applications
Imbalanced Data Classification Techniques
Network Security and Intrusion Detection
Acceso en línea:Enlace permanente (DOI o Handle)
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Sumario:Financial institutions must decide, within the milliseconds of a payment authorisation window and on evidence that is almost entirely unlabelled, whether a transaction is fraudulent: positives account for less than 0.2% of records, confirmatory labels arrive months later as chargebacks, and the generating process adapts to whatever boundary a detector exposes. Supervised classification is poorly posed under these conditions. An autoencoder is a neural network trained to reconstruct its own input through a low-dimensional bottleneck; fitted to legitimate transactions alone, it converts an intractable labelling problem into a tractable one-class fitting problem, and the reconstruction error of a new record becomes a continuous score of atypicality. This review synthesises 85 Scopus-indexed records published between 2018 and 2026. It describes eleven architectural families and their hybridisation with resampling schemes, tree ensembles, one-class classifiers, and generative models across card payments, transaction monitoring, accounting and audit records, insurance claims, anti-money laundering screening, and on-chain value transfers. It then examines how these detectors are evaluated and finds the design space broad in architecture and narrow in evaluation: threshold selection, temporally held-out validation, drift analysis, and latency reporting are absent from almost every record examined. Closing those gaps, rather than adding architectural variants, is the priority the literature indicates.

Ejemplares similares