Highlighting key portions of text within a document

• US 10,606,959 B2
• Filed: 11/20/2018
• Issued: 03/31/2020
• Est. Priority Date: 11/17/2017
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Inc.
• Inventors: Dockhorn, Carl Iwan, Fitzgerald, Sean Michael, Malangully, Ragunandan Rao, Byrum, Laurie Marie, Waters, Jason Guthrie, Thevenet, Frederic Claude, Chang, Walter Wei-Tuh
• Primary Examiner(s): Ries, Laurie A
• Application Number:
  US16/196,859
• Publication Number:
  US 20190155910A1
• Time in Patent Office: 497 Days
• CPC Class Codes: G06F 16/345 Summarisation for human users
• CPC Class Codes: G06F 40/106 Display of layout of documents; Previewing
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/47 Machine-assisted translation, e.g. using translation memory
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/025 Extracting rules from data
• CPC Class Codes: G06V 30/416 Extracting the logical structure, e.g. chapters, sections or page numbers; Identifying elements of the document, e.g. authors

Abstract

Highlighting key portions of text within a document is described. A document having text is obtained, and key portions of the document are determined using summarization techniques. Key portion data indicative of the key portions is generated and maintained for output to generate a highlighted document in which highlight overlays are displayed over or proximate the determined key portions of the text within the document. In one or more implementations, reader interactions with the highlighted document are monitored to generate reader feedback data. The reader feedback data may then be combined with the output of the summarization techniques in order to adjust the determined key portions. In some cases, the reader feedback data may also be used to improve the summarization techniques.

Claims

1. In a digital medium environment to highlight key portions of text within documents, a method implemented by a client device, the method comprising:

displaying, by the client device, a highlighted document comprising text and highlight overlays over or proximate to key portions of text within the highlighted document that collectively form a summary of the highlighted document;

receiving, by the client device, user input to modify the highlighted document, the user input received via user interaction with the highlighted document;

modifying the highlighted document based on the user input by removing at least one highlight overlay or adding at least one highlight overlay; and

displaying the modified highlighted document.

2. A method as described in claim 1, wherein the user input to modify the highlighted document is received via the user interaction with a selectable feedback control displayed with the highlighted document.

3. A method as described in claim 2, wherein the selectable feedback control is displayed proximate a respective highlight overlay and enables user input to provide positive or negative feedback regarding the respective highlight overlay.

4. The method as described in claim 3, wherein the modifying the highlighted document further comprises modifying the highlighted document by removing the respective highlight overlay in response to receiving user input to provide negative feedback via the selectable feedback control.

5. The method as described in claim 4, further comprising generating negative reader feedback data regarding the respective highlight overlay in response to receiving the user input to provide the negative feedback via the selectable feedback control, and communicating the negative reader feedback data to a service provider.

6. The method as described in claim 3, further comprising generating positive reader feedback data regarding the respective highlight overlay in response to receiving user input to provide positive feedback via the selectable feedback control, and communicating the positive reader feedback data to a service provider.

7. The method as described in claim 2, wherein the selectable feedback control is displayed in response to a user selection of un-highlighted text within the highlighted document.

8. The method as described in claim 7, further comprising:

modifying the highlighted document by adding a highlight overlay over the selected un-highlighted text in response to receiving user input to provide positive feedback via the selectable feedback control;

generating positive reader feedback data regarding the selected un-highlighted text; and

communicating the positive reader feedback data to a service provider.

9. The method as described in claim 1, wherein the highlighted document includes at least two different types of highlight overlays for at least two respective key portions based on a relative importance of each respective key portion to the highlighted document.

10. The method as described in claim 1, further comprising:

receiving additional user input to filter the highlight overlays based on at least one topic of the highlighted document; and

modifying the highlighted document based on the additional user input by removing at least one highlight overlay which does not correspond to the at least one topic.

11. A method as described in claim 1, further comprising:

receiving additional user input comprising a query;

processing the query using natural language processing techniques to determine at least one topic of the query; and

modifying the highlighted document based on the additional user input by removing at least one highlight overlay which does not correspond to the at least one topic of the query.

12. The method as described in claim 1, further comprising displaying at least one user interface element to indicate relative amounts of time users spend reading various areas of the highlighted document.

13. The method as described in claim 1, wherein the key portions of the highlighted document are determined using machine learning summarization techniques.

14. The method as described in claim 1, wherein the user interaction with the highlighted document comprises a selection of text within the highlighted document.

15. The method as described in claim 1, wherein the user interaction with the highlighted document comprises an interaction other than input to a selectable feedback control.

16. The method as described in claim 1, wherein the receiving the user input comprises at least one of tracking cursor or scrolling movement, tracking selection of text in the highlighted document, tracking a gaze of the user while reading the highlighted document.

17. A document highlighting system implemented in a digital medium environment, the document highlighting system comprising:

at least one processor;

a memory having stored thereon computer-readable instructions that are executable by the at least one processor to perform operations comprising;

obtaining a plurality of documents having text;

determining key portions of the text of the documents using a document summarization model, the key portions of the text forming a summary of the respective document;

generating highlighted versions of the documents, the highlighted versions of the document configured to display highlight overlays over the key portions of the text forming the summary;

providing the highlighted versions of the documents to client devices;

obtaining reader feedback data from the client devices based on reader interactions with the highlighted versions of documents; and

updating the document summarization model using machine learning techniques based on the reader feedback data.

18. The system as described in claim 17, wherein the reader feedback data includes positive reader feedback regarding highlight overlays of the highlighted versions of the documents and negative reader feedback regarding highlight overlays of the highlighted versions of the documents.

19. The system as described in claim 17, further comprising adjusting the key portions of an individual document based on reader feedback data associated with the individual document, wherein the adjusting includes adding or removing key portions from the highlighted version of the individual document.

20. A computing device comprising:

a display device;

at least one processor; and

a memory having stored thereon computer-readable instructions that are executable by the at least one processor to perform operations comprising;

displaying, on the display device, a highlighted document comprising text and highlight overlays over or proximate to key portions of text within the highlighted document that collectively form a summary of the highlighted document;

receiving user input to modify the highlighted document, the user input received via user interaction with the highlighted document;

modifying the highlighted document based on the user input by removing at least one highlight overlay or adding at least one highlight overlay; and

displaying the modified highlighted document on the display device.