Calculation apparatus, calculation method, learning apparatus, learning method, and program

An apparatus for learning parameters of a calculation apparatus for predicting intransitive choices from a discrete choice model, the learning apparatus comprising:

• a learning data acquisition unit configured to acquire learning data that includes a feature vector of each of a plurality of alternatives, and choice information that represents an alternative chosen by a user from the plurality of alternatives;
• and
• a learning processor configured to improve a relative evaluation of the alternative represented by the choice information by adjusting a conversion parameter that converts the feature vector into an absolute evaluation vector.

A computer-implemented method for learning parameters for a calculation method for predicting intransitive choices from a discrete choice model, the method executed by the computer comprising the steps of:

• acquiring learning data for each set of a plurality of alternatives, the learning data including a feature vector of each of the plurality of alternatives, and choice information that represents an alternative chosen by a user from the plurality of alternatives;
• and
• improving the relative evaluation of the alternative indicated by the choice information in the relative evaluation vector by adjusting a conversion parameter that converts the feature vector into the absolute evaluation vector.