FEATURE SELECTION SYSTEM, FEATURE SELECTION METHOD, AND FEATURE SELECTION PROGRAM

摘要

A feature selection system 90 that selects features to be used in learning for a predictive model when a manipulated variable for optimizing an objective function under constraint conditions is calculated, said objective function being expressed using said predictive model, said feature selection system comprising: a reception unit 91 that receives the designation of a target of prediction and the designation of a manipulated variable; a feature selection unit 92 that selects a first feature set and a second feature set from a set of features that can influence the target of prediction, said first feature set being a set of features that influence the target of prediction, said second feature set being a set of features that influence the manipulated variable; and an output unit 93 that outputs the first feature set and the second feature set.