Censoring
A common statistical condition in survival analysis where the true time-to-event for a specific observation is completely unknown or only partially known.
Think of It Like This
Like tracking how long lightbulbs last, but the study ends while five of them are still shining, so you only know they lasted 'at least' that long.
If a clinical trial ends before a patient experiences the target event, or if a user uninstalls an app before cancelling their subscription, the data is considered right-censored. Ignoring censored data and only training on completed events violently biases a model toward shorter survival times. Specialized algorithms like the Cox Proportional Hazards model explicitly account for this missing information.