ANOMALY DETECTION-AIDED AI CLASSIFIER TRAINING

Abstract

Systems and methods include use of anomaly detection rules may be initially used to detect anomalies in received data instances and to label the instances as anomalous or not anomalous. Once a sufficiently-large set of labeled data instances is available, the labeled data instances are used to train a classification model. The trained classification model and the anomaly detection rules are used to label received data instances as anomalous or not anomalous. The model is re-trained periodically using received labeled data instances until its performance exceeds a threshold. At this point, the trained model only is used to label subsequently-received data instances the instances as anomalous or not anomalous.

CPC Classifications