Detecting Anomalies Through Contrast in Heterogeneous Data
This work addresses the detection of illegal timber trade, a domain-specific problem with ecological and economic impacts, but is incremental as it builds on prior anomaly detection methods with a hybrid approach for heterogeneous data.
The paper tackles the problem of detecting fraudulent timber trade transactions by formulating it as unsupervised anomaly detection on heterogeneous data with categorical and continuous features, proposing a novel model that uses an asymmetric autoencoder and negative sampling to approximate data likelihood without assumptions about latent space structure.
Detecting anomalies has been a fundamental approach in detecting potentially fraudulent activities. Tasked with detection of illegal timber trade that threatens ecosystems and economies and association with other illegal activities, we formulate our problem as one of anomaly detection. Among other challenges annotations are unavailable for our large-scale trade data with heterogeneous features (categorical and continuous), that can assist in building automated systems to detect fraudulent transactions. Modelling the task as unsupervised anomaly detection, we propose a novel model Contrastive Learning based Heterogeneous Anomaly Detector to address shortcomings of prior models. Our model uses an asymmetric autoencoder that can effectively handle large arity categorical variables, but avoids assumptions about structure of data in low-dimensional latent space and is robust to changes to hyper-parameters. The likelihood of data is approximated through an estimator network, which is jointly trained with the autoencoder,using negative sampling. Further the details and intuition for an effective negative sample generation approach for heterogeneous data are outlined. We provide a qualitative study to showcase the effectiveness of our model in detecting anomalies in timber trade.