Cost-sensitive Active and Proactive Learning of Decision Trees - Data Mining with Decision Trees: Theory and Applications - page 201

if inferior in terms of accuracy, allows us to evaluate the consequences of
an important action: changing a price plan.
Another alternative for a supervised learning algorithm is to design an
algorithm that will enable us to find better changes in the second phase,
that is, to design an algorithm that is sensitive to the utility function and
not to accuracy. Dahan Dahan et al . (2014) introduce a full implementation
of proactive data mining for decision trees. We refer the readers to [ Dahan
et al . (2014) ] for all the details.