In the context of demonstrating freedom from infection, confidence is the probability that the type of surveillance applied would detect the presence of infection if the population were infected. The confidence depends on, among other parameters, the assumed level of infection in an infected population. The term refers to confidence in the ability of the surveillance applied to detect disease, and is equivalent to the sensitivity of the surveillance system.
Confidence of an Association Rule (X -- Y), quantifies how often and occur together as a fraction of the number of records in which occurs. For example, if the confidence is 50%, and occur together in 50% of the records in which occurs. Thus, knowing that occurs in a record, the probability that also occurs in that record is 50%.
Estimates of future performance always entail a confidence factor, the inverse of uncertainty. It is useful to quantify the confidence to perform Sensitivity Analysis, demonstrating the possible impacts of different levels of performance. See Sensitivity Analysis.