US 12,130,850 B2
Model-based semantic text searching
Trung Bui, San Jose, CA (US); Yu Gong, San Jose, CA (US); Tushar Dublish, Uttar Pradesh (IN); Sasha Spala, Arlington, MA (US); Sachin Soni, New Delhi (IN); Nicholas Miller, Boston, MA (US); Joon Kim, Dublin, CA (US); Franck Dernoncourt, Sunnyvale, CA (US); Carl Dockhorn, San Jose, CA (US); and Ajinkya Kale, San Jose, CA (US)
Assigned to Adobe Inc., San Jose, CA (US)
Filed by Adobe Inc., San Jose, CA (US)
Filed on Dec. 29, 2022, as Appl. No. 18/147,960.
Application 18/147,960 is a continuation of application No. 16/849,885, filed on Apr. 15, 2020, granted, now 11,567,981.
Prior Publication US 2023/0133583 A1, May 4, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/33 (2019.01); G06F 40/30 (2020.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01)
CPC G06F 16/3347 (2019.01) [G06F 40/30 (2020.01); G06N 5/04 (2013.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
generating, on an end-user device, a set of tokens from a portion of an electronic document in response to accessing the electronic document, wherein the portion of the electronic document is a subset of the electronic document and each token corresponds to one or more strings within the electronic document;
monitoring a user interface of the end-user device for one or more characters of an input, wherein the input corresponds to a request to locate text within the electronic document that is associated with a keyword included in the input;
responsive to determining that a condition associated with the monitored one or more characters of the input is satisfied, sending a number of characters of the monitored one or more characters of the input and the set of tokens to a machine learning system configured to determine a semantic mapping based on a similarity between the monitored one or more characters of the input and the set of tokens;
in response to detecting input of a new character of the keyword, after the keyword and the set of tokens are sent for semantic mapping, sending an updated keyword, including the new character, to the machine learning system;
receiving, from the machine learning system, at least one string within the portion of the electronic document that is semantically related at least to the monitored one or more characters of the input; and
outputting, on the user interface of the end-user device, an indication of the at least one string semantically related to the monitored one or more characters of the input.