Provided is a classification tree generation device 10 for selecting, from a plurality of classification condition candidates, a new classification condition to be added to a classification tree, which is a prediction model expressed in a tree structure formed from one or more nodes representing classification conditions, said device comprising: a first calculation part 11 for calculating information gain relating to the classification condition candidates; a second calculation part 12 for calculating, as a cost relating to the classification condition candidates, a value representing the magnitude of the smallest difference among differences between the classification condition candidates and each of the classification conditions included in the classification tree; and a selection part 13 for selecting, as the new classification condition, the classification condition candidate from among the plurality of classification condition candidates that has the largest value among values obtained by subtracting the calculated cost from the calculated information gain.
展开▼