Classification system
View Patent ↗A classification system includes classification training data-specific classification data correct attribute setting configured to set not only the correct classes of pieces of classification training data but also the correct attributes of the pieces of classification training data and classification/attribute training configured to train classification by comparing the class/attribute estimation result outputs with the correct classes set by correct class setting and comparing the attribute estimation result outputs with the correct attributes set by the classification training data-specific classification training data correct attribute setting. Pieces of classification target data from classification target data acquisition are newly inputted to the trained classification means. Classification/attribute result output outputs both the classification results and attribute results of the pieces of classification target data so that validity of the classification results can be considered.
1 . A classification system comprising a processor, wherein the processor:
obtains pieces of classification training data;
sets correct classes with respect to the pieces of classification training data;
receives the pieces of classification training data and outputs classification results;
trains the processor by comparing the classification results with the correct classes previously set and feeding back an error,
sets not only the correct classes of the pieces of classification training data but also correct attributes of the pieces of classification training data,
receives the pieces of classification training data; and
outputs not only the classification results but also class/attribute estimation results including attribute estimation results,
compares the class/attribute estimation results outputted from the processor with the correct classes and the correct attributes set and, if both are not matched, make an adjustment so that the class/attribute estimation results come close to the correct classes and the correct attributes by feeding back an error to the processor,
wherein
pieces of classification target data are newly inputted to the trained processor and
the processor outputs both classification results and attribute estimation results of the pieces of classification target data so that validity of the classification results can be considered,
wherein the processor:
manages attribute lists corresponding to the classes; and
estimates more likely classes of the inputted pieces of classification target data by applying the pieces of classification target data to the trained processor and checking obtained attribute lists against the attribute lists managed.
2 . The classification system of claim 1 , wherein the processor if the obtained attribute lists include an attribute that is not included in the attribute lists managed, makes an inquiry the attribute.
3 . The classification system of claim 1 , wherein
the trained processor comprises a plurality of trained processor disposed in parallel, and
the processor estimates correct classes of the pieces of classification target data by checking, against the managed attribute lists, attribute lists obtained by applying the pieces of classification target data to the plurality of trained processor disposed in parallel.