Anomaly Detection is an unsupervised Machine Learning task which identifies instances in a dataset that do not conform to a regular pattern. BigML anomaly detector (called anomaly in BigML) is an optimized implementation of the Isolation Forest algorithm that is capable to detect anomalous instances and suspicious patterns in unlabeled datasets given a set of input fields. This means that you do not need to collect a training dataset knowing in advanced which instances are anomalous and which are normal. Anomaly Detection is used for fraud detection, data cleansing tasks, predictive maintenance and intrusion detection, among other examples.
Watch this video to learn how to identify unusual instances in your data using BigML Anomaly Detector through the BigML Dashboard:
For more details, please find the documentation about anomalies here if you are using the Dashboard, or here if you prefer the API.