Evaluation & Quality
Specificity
How often a system correctly leaves cases unflagged when the condition it looks for is absent.
Example
A detector correctly leaves 95 of 100 non-defective items unflagged.
Why people use it
It shows how often a detector correctly leaves ordinary cases unflagged.
What you'll hear
“How often does it correctly leave safe items alone?”
What this means for you
Consider specificity together with recall and the costs of both error types.
Can you control it?
No
No direct control. This describes a wider issue, concept or result rather than something you can simply switch on or off in a tool.
Common questions
- Can a rare condition make false alarms especially noticeable?
- Yes. Even a small false-alarm rate can produce many alerts when most checked cases do not have the condition.
- Can a detector get perfect specificity by flagging nothing?
- Yes. It would also miss every genuine positive case, which shows why another measure is needed.
- Does the score explain the cost of false alarms?
- No. The same false-alarm rate can have very different practical consequences.