Anomaly detection

In the Ultra Transcenders books

AI-901PL-300

Each book explains Anomaly detection in context, with comparison tables and the common traps.

Terms in this definition

• VALUES

  Returns in DAX the distinct column values, or table rows, still visible after filters are applied, sometimes with an extra blank entry. CALCULATE often takes the result as a table filter.

• Dedicated

  Running Azure Functions on an App Service plan, which removes the execution time limit and offers VNet integration on Basic and higher tiers.

Related terms

• Advanced Analytics

  Adds device query, device scopes, a device timeline, anomaly detection, resource performance and battery health on top of endpoint analytics. It comes with the Intune Suite and with Intune Plan 2.

• Analytics pane

  A tab beside the formatting options in the Visualizations pane for adding extras such as dynamic reference lines, error bars, forecasting and anomaly detection. Which of these appear varies with the kind of visual selected.

• Azure Stream Analytics

  Serverless service that processes streaming data with a SQL-like language, offering windowing functions and anomaly detection out of the box.

• Data profiling

  One part of data quality monitoring. Once a table has a profile attached, statistics and drift are calculated for it over time, with results saved to tables and shown on a dashboard; anomaly detection is the equivalent that covers an entire schema.

• Dynamic threat detection model

  Since June 2025, anomaly detection in Defender for Cloud Apps runs on this adaptive model. Moving to it meant switching off a number of older anomaly policies and giving their detections new names.

• Population Z-score model

  An anomaly detection model in Endpoint analytics that picks out unusual devices or apps using a dataset's mean and standard deviation; it is only accurate with large amounts of data.

• Prevalence rate

  In Endpoint analytics anomaly detection, the proportion of a group's devices that an anomaly affects.