Confidence within data mining is typically utilized for association-rule learning. In the case of market-basket analysis, confidence describes the relationships between two products. Namely, when an individual has item x in their shopping cart, what is the likelihood that this item is accompanied by item y? A further inference type may be taken with additional data along the lines of if an individual is buying onions and potatoes, they may be making burgers and fries.
Association-rule learning has applications much beyond market-basket analysis, and is utilized across large data sets to gain insights within intrusion detection (information assurance), web usage data, bioinformatics, and continuous production methodologies.