NORMALITY/ABNORMALITY DISCRIMINATION DEVICE, NORMALITY/ABNORMALITY DISCRIMINATION METHOD, AND NORMALITY/ABNORMALITY DISCRIMINATION SYSTEM

摘要

PROBLEM TO BE SOLVED: To promptly and appropriately perform discrimination of whether a discrimination object is normal or abnormal even when the amount of measured data for a discrimination object is small.SOLUTION: A computing machine 1 comprises a CPU 11 which: specifies a first learning category for classifying whether measured data is data indicating normality or data indicating abnormality based on reference data; specifies a second learning category for classifying whether measured data is data indicating normality or data indicating abnormality based on standard data; specifies an arrangement of corresponding learning categories corresponding to non-corresponding learning categories for which the corresponding second learning category does not exist to the measured data based on a positional relationship between the first learning category and the corresponding second learning category; and discriminates whether a discrimination object is normal or abnormal depending on the association of discrimination object data with the second learning category and the corresponding learning category.SELECTED DRAWING: Figure 1