Model generation system and model generation method

A model generation system, comprising:

• a NCLM processor narrowing down a first input variable group to a second input variable group selected using the Nearest Correlation Louvain Method (NCLM), the first input variable group including a plurality of input variables;
• a filter part narrowing down the second input variable group to a third input variable group satisfying a prescribed condition;
• a model generator generating a model of a relationship between the third input variable group and an output variable;
• a variable narrow-down part narrowing down the second input variable group to one or more of the input variables not used in the generation of the model;
• a determiner determining whether or not the number of the generated models has reached a specified number, and in the case where the number of the models has not reached the specified number, the determiner outputting to the filter part the second input variable group narrowed down by the variable narrow-down part;
• and
• a soundness calculator calculating an overall soundness of the specified number of the models and calculating a soundness of each of the models.

A model generation system, comprising:

• a NCLM processor narrowing down a first input variable group to a second input variable group selected using the Nearest Correlation Louvain Method (NCLM), the first input variable group including a plurality of input variables;
• a filter part narrowing down the second input variable group to a third input variable group satisfying a prescribed condition;
• a model generator generating a model of a relationship between the third input variable group and an output variable;
• a variable narrow-down part narrowing down the first input variable group to one or more of the input variables not used in the generation of the model;
• a determiner determining whether or not the number of the generated models has reached a specified number, and in the case where the number of the models has not reached the specified number, the determiner outputting, as the second input variable group to the filter part, the first input variable group narrowed down by the variable narrow-down part;
• and
• a soundness calculator calculating an overall soundness of the specified number of the models and calculating a soundness of each of the models.