Multi-label classification using a learned combination of base classifiers

摘要

Multi-label classification is performed by (i) applying a set of trained base classifiers to an object to generate base classifier label prediction sets comprising subsets of a set of labels; (ii) constructing a set of second level features including at least one second level feature defined by a predetermined combination of two or more of the base classifier label prediction sets; and (iii) applying a second level classifier to label the object with a set of one or more labels comprising a subset of the set of labels, labeling being based on the set of second level features. The multi-label classifier is trained by: (iv) applying operations (i) and (ii) to labeled training objects of a set of labeled training objects to generate training metadata comprising sets of second level features for the labeled training objects; and (v) training the second level classifier using the training metadata.

主权项

1. A method comprising:
performing multi-label classification of an object respective to a set of labels by operations including:
(i) applying a set of trained base classifiers to the object to generate base classifier label prediction sets comprising subsets of the set of labels wherein at least two base classifiers of the set of trained base classifiers are multi-label classifiers;(ii) generating a set of second level features each comprising a predetermined combination of one or more of the base classifier label prediction sets wherein at least two second level features of the set of second level features each comprise a predetermined combination of two or more of the base classifier label prediction sets constructed using one or more set operators;(iii) applying a second-level multi-label classifier that receives input data comprising the set of second level features and based on the input data generates as output a set of one or more labels comprising a subset of the set of labels; wherein the multi-label classification is performed by a digital processing device.