Hierarchical classification-based incremental class learning method and computing device for digital storytelling

摘要

The present invention relates to a hierarchical classification based incremental class learning method and computing device for digital storytelling that proposes a hierarchical classification model capable of incremental class learning when recommending an appropriate multimedia or service. Such an aspect of the present invention is an incremental class learning method that enables hierarchical classification using a classification module composed of a blocking ARTMAP module and a classification ARTMAP module, (a) input information of different channels is classified with the blocking ARTMAP module Input to each ARTMAP module; (b) The input vector reflecting the input information of the different channels input in step (a) is input to the blocking ARTMAP module of any layer, and the incremental information to the next layer is output, and in step (a) Inputting the input vector reflecting input information of different channels into a classification ARTMAP module of any layer, where the class vector of the arbitrary layer is output; And (c) adding a class vector of an arbitrary layer output in step (b) to an input vector reflecting input information of different channels in step (b).