Programming ML.NET

Chapter 7

Anomaly Detection Tasks

“Truths of this kind should be drawn from notions rather than from notations”.

—Carl Friedrich Gauss, Disquisitiones Arithmeticae, 1801

In data mining and machine learning, anomaly detection refers to the steps necessary to identify unusual occurrences of items in an amount of data that might be quite large. At the same time, many high-level business problems can be formulated as instances of the core anomaly detection problem, such as detecting bots in web pages, spotting outliers in sales reports, flagging suspicious fraudulent behaviors, and monitoring the health of industrial machines. Furthermore, in machine learning, anomaly detection is also used to remove outliers from a dataset to augment the accuracy ...