| 摘要 |
PROBLEM TO BE SOLVED: To automatically estimate a suitable model when data is provided, and to learn and output the estimated model.SOLUTION: In the invention, model generation means of model selection means generates a new model set on the basis of a model set and a calculated cost that are input from cost calculation means. Model learning means learns each model on the basis of the new model set input from the model generation means and data stored in data storage means, and outputs the learned model set. The cost calculation means iterates processing of calculating the cost for each model, on the basis of the learned model set input from the model learning means. |
