| CPC G06F 3/0482 (2013.01) [G06F 3/04847 (2013.01); G06F 16/532 (2019.01); G06F 16/5846 (2019.01); G06F 16/5866 (2019.01); G06F 2203/04806 (2013.01)] | 18 Claims |
|
1. A method, comprising:
accessing a multi-element document, the multi-element document including a plurality of elements depicted in the multi-element document;
determining a global intent for the multi-element document based on the plurality of elements depicted in the multi-element document, wherein determining the global intent for the multi-element document comprises:
determining, for each of the plurality of elements, an element-specific topic distribution comprising a ranked list of topics;
wherein at least one of the plurality of elements comprises a text element having less than a threshold length, and wherein determining the element-specific topic distribution for each text element having less than the threshold length comprises applying a machine learning model to text associated with the text element to generate the ranked list of topics specific to the text element; and
creating a first aggregated topic distribution from the determined element-specific topic distributions, wherein the global intent comprises one or more terms from the first aggregated topic distribution;
selecting, from the plurality of elements, a selected element;
generating, based on the determined global intent, a replacement multi-element document that includes a substitute element in place of the selected element in the multi-element document, wherein content of the substitute element comprises one or more features that are different from one or more corresponding features of content of the selected element; and
displaying, via a user interface with the multi-element document, a preview of the replacement multi-element document providing a view of the replacement multi-element document, wherein the view of the replacement multi-element document is focused to depict the substitute element.
|