US 10,891,430 B2
Semi-automated methods for translating structured document content to chat-based interaction
Scott Carter, Menlo Park, CA (US); Laurent Denoue, Vento (IT); and Matthew L. Cooper, San Francisco, CA (US)
Assigned to FUJI XEROX CO., LTD., Tokyo (JP)
Filed by FUJI XEROX CO., LTD., Tokyo (JP)
Filed on Feb. 2, 2018, as Appl. No. 15/887,927.
Prior Publication US 2019/0243889 A1, Aug. 8, 2019
Int. Cl. G06F 40/174 (2020.01); G06F 3/0488 (2013.01); G06K 9/00 (2006.01); G06F 16/332 (2019.01); G06F 40/103 (2020.01)
CPC G06F 40/174 (2020.01) [G06F 3/04883 (2013.01); G06F 16/3329 (2019.01); G06F 40/103 (2020.01); G06K 9/00469 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A method of converting a document from a first structure to a second structure, comprising:
extracting data of the document to associate fields and labels in the first structure to generate field/label associations of the document for the first structure;
generating a grouping comprising a field/label association by an automatic input that does not require an operator;
receiving a non-automated operator input selecting the field/label association and one or more other field/label associations or an open label;
updating the grouping, in a semi-automatic manner, based on the received non-automated operator input to associate the field/label association with the one or more other field/label associations or the open label; and
based on the operator input and a spatial arrangement of the first structure, providing the grouping in the second structure, the grouping is remapped as a natural conversational unit that replaces the field and the label,
wherein for the operator input provided at the first structure, and the natural conversation unit comprises a single query including the field/label association and the one or more other field/label associations or the open label,
wherein the operator input comprises at least one of a gesture and a line that remaps the field and label of the first structure into the natural conversational unit of the second structure, and
wherein the gesture is a mark associated with the received operator input, wherein the mark contacts the field/label association to be grouped with the one or more other field/label associations.