IP Library Granted Patent US 12,579,519
Granted Patent B2
US 12,579,519 · App. 17/583,767 · Granted Mar 17, 2026

Generating digital associations between documents and digital calendar events based on content connections

Inventors: Kavitha Radhakrishnan (San Jose, CA); Timo Mertens (Millbrae, CA); Neil Sethi (San Francisco, CA); David Stinnette (San Francisco, CA); Justin Hileman (San Francisco, CA); Adam LaFave (San Francisco, CA)
Assignee: Dropbox, Inc.
G06Q10/1093G06F16/93G06N20/00H04L63/102H04L65/403
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,579,519
App. No.
17/583,767
Granted
Mar 17, 2026
Kind
B2
Abstract

The present disclosure relates to generating and utilizing digital associations between digital documents and digital calendar events based on determined content connections. For example, one or more embodiments suggest a digital document to a user for association with a digital calendar event based on content connections between the digital document and the digital calendar event. In particular, one or more embodiments determine a content connection based on document characteristics (e.g., document content and access characteristics) and calendar event characteristics (e.g., calendar content and historical meeting information). For example, the system can train a machine learning model to determine content connections based on document characteristics and calendar event characteristics. By associating digital documents with digital calendar events, the system can efficiently manage digital documents for participants, automatically generate new digital documents for calendar events, generate snippets for insertion within digital documents, and generate calendar event summaries and action items.

Claims (55)

1 . A method, comprising:

identifying a first digital calendar event associated with a user account;

determining a content connection between the first digital calendar event and a digital document of a plurality of digital documents associated with the user account by processing document content of the digital document and calendar content of the first digital calendar event, wherein determining the content connection comprises determining the content connection using a machine learning model trained by having parameters iteratively updated using feedback determined from comparing predicted content connections generated by the machine learning model during training to corresponding ground truths;

identifying a second digital calendar event associated with the user account, wherein the first digital calendar event and the second digital calendar event are within a predetermined time period;

generating a first calendar event indicator corresponding to the first digital calendar event and a second calendar event indicator corresponding to the second digital calendar event, wherein the first calendar event indicator comprises a link to the first digital calendar event and an additional link to the digital document having the content connection with the first digital calendar event; and

modifying a user interface of a computing device associated with the user account to include the first calendar event indicator and the second calendar event indicator.

2 . The method of claim 1 , further comprising:

receiving an indication of a user interaction with the additional link to the digital document; and

providing the computing device access to the digital document within a document application.

3 . The method of claim 1 , wherein determining the content connection comprises comparing the document content with the calendar content, wherein the document content comprises at least one of a document title or document text, and the calendar content comprises at least one of a digital calendar event title or a digital calendar event description.

4 . The method of claim 1 , wherein the first calendar event indicator further comprises a snippet of the digital document.

5 . The method of claim 1 , further comprising:

determining at least one additional content connection between at least one additional digital document and the first digital calendar event,

wherein the first calendar event indicator comprises a list of digital documents comprising the digital document and the at least one additional digital document.

6 . The method of claim 1 , wherein:

identifying the first digital calendar event and the second digital calendar event comprises identifying upcoming digital calendar events for the user account; and

modifying the user interface of the computing device to include the first calendar event indicator and the second calendar event indicator comprises modifying the user interface before the first digital calendar event and the second digital calendar event occur.

7 . The method of claim 1 , further comprising generating a summary document for the digital document by analyzing the document content of the digital document to identify key sentences of the document content by:

generating an intersection score for each sentence pair from sentences of a paragraph of the document content;

determining a score for each sentence from the paragraph based on intersection scores for sentence pairs that include the sentence; and

extracting at least one sentence from the paragraph as a key sentence based on the score for each sentence from the paragraph.

8 . The method of claim 7 , wherein generating the intersection score for each sentence pair from the sentences of the paragraph of the document content comprises:

dividing each of a first sentence and a second sentence included in the sentence pair into words;

determining a number of words that are common between the first sentence and the second sentence included in the sentence pair; and

generating the intersection score by dividing the number of words that are common by an average length of the first sentence and the second sentence.

9 . A non-transitory computer readable storage medium comprising instructions that, when executed by at least one processor, cause a computer system to:

identify a first digital calendar event associated with a user account;

determine a first content connection between the first digital calendar event and a first digital document of a plurality of digital documents associated with the user account, wherein determining the first content connection comprises determining the first content connection using a machine learning model trained by having parameters iteratively updated using feedback determined from comparing predicted content connections generated by the machine learning model during training to corresponding ground truths;

identify a second digital calendar event associated with the user account, wherein the first digital calendar event and the second digital calendar event are within a predetermined time period;

generate a first calendar event indicator corresponding to the first digital calendar event and a second calendar event indicator corresponding to the second digital calendar event, wherein the first calendar event indicator comprises a link to the first digital calendar event and an additional link to the summary document for the first digital document having the first content connection with the first digital calendar event; and

modify a user interface of a computing device associated with the user account to include the first calendar event indicator and the second calendar event indicator.

10 . The non-transitory computer readable storage medium of claim 9 , wherein determining the first content connection between the first digital document and the first digital calendar event comprises utilizing a machine learning model to determine the first content connection.

11 . The non-transitory computer readable storage medium of claim 10 , wherein the instructions, when executed by the at least one processor, cause the computer system to:

determine a second content connection between a second digital document of the plurality of digital documents and the second digital calendar event; and

based on the second content connection, provide a link to the second digital document within the second calendar event indicator.

12 . The non-transitory computer readable storage medium of claim 9 , wherein the predetermined time period is a current day.

13 . The non-transitory computer readable storage medium of claim 9 , wherein the instructions, when executed by the at least one processor, cause the computer system to provide the computing device access to the first digital calendar event within a calendar application based on receiving an indication of a user interaction with the link to the first digital calendar event.

14 . The non-transitory computer readable storage medium of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the computer system to provide the computing device access to the first digital document within a document application based on receiving an indication of a user interaction with the additional link to the first digital document.

15 . A system comprising:

at least one processor; and

a non-transitory computer readable storage medium comprising instructions that, when executed by the at least one processor, cause the system to:

identify a first digital calendar event associated with a user account;

determine a content connection between the first digital calendar event and a digital document of a plurality of digital documents associated with the user account by processing document content of the digital document and calendar content of the first digital calendar event, wherein determining the content connection comprises determining the content connection using a machine learning model trained by having parameters iteratively updated using feedback determined from comparing predicted content connections generated by the machine learning model during training to corresponding ground truths;

identify a second digital calendar event associated with the user account, wherein the first digital calendar event and the second digital calendar event are within a predetermined time period;

generate a first calendar event indicator corresponding to the first digital calendar event and a second calendar event indicator corresponding to the second digital calendar event, wherein the first calendar event indicator comprises a link to the first digital calendar event and an additional link to the summary document for the digital document having the content connection with the first digital calendar event; and

modify a user interface of a computing device associated with the user account to include the first calendar event indicator and the second calendar event indicator.

16 . The system of claim 15 , wherein the first calendar event indicator and the second calendar event indicator are provided in an application that is not a calendar application.

17 . The system of claim 15 , further comprising instructions that, when executed by the at least one processor, cause the system to:

determine at least one additional content connection between at least one additional digital document and the first digital calendar event,

wherein the first calendar event indicator comprises a list of digital documents comprising the digital document and the at least one additional digital document.

18 . The system of claim 15 , wherein the first calendar event indicator further comprises a snippet of the digital document.

19 . The system of claim 15 , wherein determining the content connection between the digital document and the first digital calendar event comprises utilizing a machine learning model to determine the content connection.

20 . The system of claim 15 , further comprising instructions that, when executed by the at least one processor, cause the system to:

provide the computing device access to the first digital calendar event within a calendar application based on receiving an indication of a user interaction with the link to the first digital calendar event; and

provide the computing device access to the digital document within a document application based on receiving an indication of a user interaction with the additional link to the digital document.

Assignments (2)
SECURITY INTEREST Recorded Dec 12, 2024
From: DROPBOX, INC.
To: WILMINGTON TRUST, NATIONAL ASSOCIATION, AS COLLATERAL AGENT
Reel/Frame 069604/0611 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 25, 2022
From: RADHAKRISHNAN, KAVITHA; MERTENS, TIMO; SETHI, NEIL; STINNETTE, DAVID; HILEMAN, JUSTIN; LAFAVE, ADAM
To: DROPBOX, INC.
Reel/Frame 058761/0643 →
Continuity (2)
Continuation 15908312 · Feb 28, 2018
Related Publication 20220147948A1 · May 12, 2022
References Cited (48)
US 5899979A · Miller et al. · 1999 [cited by applicant]
US 6085205A · Peairs et al. · 2000 [cited by applicant]
US 7925528B2 · Biazetti et al. · 2011 [cited by applicant]
US 8489615B2 · Dhara et al. · 2013 [cited by applicant]
US 9015301B2 · Redlich et al. · 2015 [cited by applicant]
US 9477949B2 · Myka et al. · 2016 [cited by applicant]
US 9631930B2 · McGavran et al. · 2017 [cited by applicant]
US 9665597B2 · Haitani et al. · 2017 [cited by applicant]
US 20080033957A1 · Forstall et al. · 2008 [cited by applicant]
US 20100306265A1 · Jones, III et al. · 2010 [cited by applicant]
US 20130036117A1 · Fisher · 2013 [cited by examiner]
US 20140181219A1 · Wang et al. · 2014 [cited by applicant]
US 20150261732A1 · Kim et al. · 2015 [cited by applicant]
US 20160163007A1 · Porter et al. · 2016 [cited by applicant]
US 20160259761A1 · Laborczfalvi et al. · 2016 [cited by applicant]
US 20160294894A1 · Miller · 2016 [cited by applicant]
US 20160373456A1 · Vermeulen et al. · 2016 [cited by applicant]
US 20170063908A1 · Muddu et al. · 2017 [cited by applicant]
US 20170067747A1 · Ricci · 2017 [cited by applicant]
US 20170078322A1 · Seiver et al. · 2017 [cited by applicant]
US 20170124487A1 · Szeto et al. · 2017 [cited by applicant]
US 20170171124A1 · Brezina · 2017 [cited by examiner]
US 20170193349A1 · Jothilingam · 2017 [cited by examiner]
US 20170235848A1 · Van Dusen et al. · 2017 [cited by applicant]
US 20170330109A1 · Maughan et al. · 2017 [cited by applicant]
US 20180189733A1 · Newhouse · 2018 [cited by examiner]
US 20180189744A1 · Frank · 2018 [cited by examiner]
US 20180239959A1 · Bui et al. · 2018 [cited by applicant]
US 20180284735A1 · Cella et al. · 2018 [cited by applicant]
US 20180300324A1 · Ziraknejad et al. · 2018 [cited by applicant]
US 20180329935A1 · Mugali et al. · 2018 [cited by applicant]
US 20180337871A1 · Matta et al. · 2018 [cited by applicant]
US 20190236555A1 · Bennett · 2019 [cited by examiner]
US 20190266573A1 · Radhakrishnan · 2019 [cited by examiner]
US 20190272507A1 · Libin · 2019 [cited by examiner]
US 20190324435A1 · Cella et al. · 2019 [cited by applicant]
US 20200042949A1 · Johnson, III · 2020 [cited by examiner]
US 20200219062A1 · Radhakrishnan · 2020 [cited by examiner]
US 20200293999A1 · Bhattacharya · 2020 [cited by examiner]
US 20200372432A1 · Newhouse · 2020 [cited by examiner]
US 20210065320A1 · Bleiweiss · 2021 [cited by examiner]
US 20220045975A1 · Flöther · 2022 [cited by examiner]
Liu, Jiangwei, An Overview of Event Extraction and Its Applications , Nov. 8, 2021, ARXIV.org, https://arxiv.org/pdf/2111.03212, p. 1 -23. (Year: 2021). [cited by examiner]
Buchanan M.C., “Automatically Generating Consistent Schedules for Multimedia Documents,” 2005, Multimedia Systems, 1, pp. 55-67. [cited by applicant]
Final Office Action from U.S. Appl. No. 15/908,312, mailed Mar. 2, 2021, 28 pages. [cited by applicant]
Non-Final Office Action from U.S. Appl. No. 15/908,312, mailed Jun. 15, 2020, 18 pages. [cited by applicant]
Non-Final Office Action from U.S. Appl. No. 15/908,312, mailed Jun. 18, 2021, 14 pages. [cited by applicant]
Notice of Allowance from U.S. Appl. No. 15/908,312, mailed Oct. 7, 2021, 12 pages. [cited by applicant]