US 12,236,189 B2
Personalized text proofing using distributed services
James Aidan Cogley, Dublin (IE); Enrico Cadoni, Dublin (IE); Colin Laird, Dublin (IE); Shashank Shekhar Gupta, Dublin (IE); and Olivier Gauthier, Duvall, WA (US)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Oct. 15, 2021, as Appl. No. 17/502,933.
Prior Publication US 2023/0124709 A1, Apr. 20, 2023
Int. Cl. G06F 40/232 (2020.01); G06F 40/166 (2020.01); G06F 40/253 (2020.01)
CPC G06F 40/232 (2020.01) [G06F 40/166 (2020.01); G06F 40/253 (2020.01)] 18 Claims
OG exemplary drawing
 
1. A method comprising:
generating, based on user signals indicating past actions of a user, a user model that is used to personalize generic critiques used for text and grammar proofing a document, the generating the user model including generating mappings of personalized autocorrect words, the generating the mappings comprising:
searching for entries from a user speller map in a tenant speller map of a group of users of which the user is a member, the tenant speller map comprising an aggregation of user speller maps of the group; and
based on finding an entry from the user speller map in the tenant speller map and based on a condition being satisfied, adding a type/meant pair corresponding to the entry to the mapping of personalized autocorrect words;
during runtime of an application on a client device used to create the document, receiving, by an augmentation system that acts as an intermediate between the client device and an editor system, a user token from the client device;
using the user token, accessing and locally caching the user model at the augmentation system;
receiving, from the client device by the augmentation system, user inputs comprising typed components used to create the document;
accessing, from the editor system by the augmentation system, a set of one or more generic critiques for the user inputs, the accessing comprising calling the editor system with the user inputs received from the client device and receiving the set from the editor system in response;
applying, by one or more hardware processors of the augmentation system, the user model to the set received from the editor system, the applying comprising overriding one or more generic critiques of the set based on the user model to generate a modified set of critiques, the modified set of critiques including generic critiques and the one or more overridden generic critiques; and
transmitting, over a network by the augmentation system, the modified set of critiques to be automatically applied or suppressed to the client device, the client device applying the modified set to the document.