For Data Mining, which of the following is the definition of the training data set?

It is the set of new data with unknown classifications which we are trying to estimate

It is a partition of the data with known classifications $($i$.$e$.$ historical data$)$ which is used to 'teach' the data mining algorithm $($i$.$e$.$ the true classifications are known$/$visible to the algorithm$)$

It is the data mining term for all of the historical data $($i$.$e$.$ the data with known classifications$)$

It is a partition of the data with known classifications $($i$.$e$.$ historical data$)$ whose classifications are set aside to test how the data mining algorithm performs $($i$.$e$.$ the true classifications are not known