Skip to content
Free public beta until December 4, 2026: every course is open, no card needed. After that, the first chapter of each course stays free and full access is a one-time $39 ($19 for up to 100 founding members chosen for helpful feedback). Your progress is kept. How the beta works

anomaly score

Any number that grows as a value looks less normal, turned into a decision by a threshold.

An anomaly score is a scalar measure of how unusual an observation is relative to a model of normal behaviour, such as a z-score; raise and clear lines with a hold count turn it into an alert state.

Examples

  • The fan detector's z-score is its anomaly score; raising above z = 4 for 3 windows in a row and clearing below z = 2 turn it into a decision.

  • Any number that grows as a window looks less normal can serve, as long as lines and a hold count turn it into an alert.

Read the full section →