US 12,079,569 B2
Document editing models and localized content management
Peter Eric Loforte, Issaquah, WA (US); Raymond Robert Ringhiser, Maple Valley, WA (US); Katharine Elizabeth Grant, Seattle, WA (US); Kapil Kumar Tundwal, Kirkland, WA (US); Paul Fraedrich Estes, Bellevue, WA (US); Veronica G. Sievers, Redmond, WA (US); and Adam Dewayne Miller, North Bend, WA (US)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by MICROSOFT TECHNOLOGY LICENSING, LLC, Redmond, WA (US)
Filed on Jun. 7, 2019, as Appl. No. 16/435,069.
Prior Publication US 2020/0387567 A1, Dec. 10, 2020
Int. Cl. G06F 40/197 (2020.01); G06F 16/31 (2019.01); G06F 16/93 (2019.01); G06F 40/166 (2020.01)
CPC G06F 40/197 (2020.01) [G06F 16/313 (2019.01); G06F 16/93 (2019.01); G06F 40/166 (2020.01)] 19 Claims
OG exemplary drawing
 
1. A system for managing edits to localized versions of electronic content, the system comprising:
a processor; and
machine-readable media including instructions which, when executed by the processor, cause the processor to:
determine that a first edit event has occurred during access of a first localized version of a first base electronic content, the first base electronic content comprising electronic content in a first language, and the first localized version of the first base electronic content comprising at least a portion of the first base electronic content translated to a second language, the first edit event including a first modification of a first portion of the first localized version by a first user;
generate a first score for the first edit event based on at least a first content characteristic of the first modification and a first user characteristic that describes a usage history of the first user;
determine that the first modification has a high likelihood of being substantive based on at least the first score using a model which predicts whether modifications are substantive;
initiate, in response to the determination that the first modification has the high likelihood of being substantive, an automated propagation of the first modification to at least the first base electronic content;
present a message to a document manager requesting confirmation that the first modification was indeed substantive; and
reinforce the model using the confirmation from the document manager,
wherein the model determines whether the modifications are substantive by comparing scores of edit events with a first threshold and a second threshold such that the edit events having a score that falls below the first threshold are non-substantive edits, the edit events having a score that exceeds the first threshold and falls below the second threshold are ambiguous edits and the message to the document manager is presented, and the edit events having a score that exceeds the first threshold and the second threshold are substantive edits which are propagated without requiring confirmation from the document manager.