US 11,809,450 B2
Selectively identifying and recommending digital content items for synchronization
Ermo Wei, Centreville, VA (US); Jialiang Li, San Francisco, CA (US); Kaiyue Sun, San Francisco, CA (US); Li Chen Koh, San Francisco, CA (US); Mingye Xia, San Francisco, CA (US); Yu Zhang, San Jose, CA (US); and Yuyang Guo, Mountain View, CA (US)
Assigned to Dropbox, Inc., San Francisco, CA (US)
Filed by Dropbox, Inc., San Francisco, CA (US)
Filed on May 16, 2022, as Appl. No. 17/663,628.
Application 17/663,628 is a continuation of application No. 15/964,267, filed on Apr. 27, 2018, granted, now 11,334,596.
Prior Publication US 2022/0277020 A1, Sep. 1, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/27 (2019.01); G06F 16/178 (2019.01); G06F 3/0482 (2013.01); H04L 67/50 (2022.01)
CPC G06F 16/27 (2019.01) [G06F 16/178 (2019.01); H04L 67/535 (2022.05); G06F 3/0482 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
identifying a user account and a set of additional user accounts having joint access to a shared collection of content items stored within a content management system;
generating user recall data comprising an indication of a number of user edits based on the user account editing one or more content items from the shared collection of content items, wherein the user recall data further comprises user synchronization data that indicates whether each content item of the shared collection of content items is synchronized to a client device associated with the user account;
generating team recall data comprising an indication of a number of team edits based on one or more accounts of the set of additional user accounts editing the contents of one or more content items from the shared collection of content items, wherein the team recall data further comprises team synchronization data that indicates whether each content item of the shared collection of content items is synchronized to client devices associated with the set of additional user accounts;
processing the user recall data and the team recall data through a machine learning model to determine, for the user account, a synchronization score with respect to each content item of the shared collection of content items, the synchronization score for a given content item representing a probability that the user account will access the given content item; and
providing a set of content items from the shared collection of content items to the client device associated with the user account based on determining that synchronization scores for the set of content items satisfy a first synchronization criteria.