IP Library Granted Patent US 10,986,046
Granted Patent B2
US 10,986,046 · App. 16/555,297 · Granted Apr 20, 2021

Apparatus and method for generating summary of conversation storing

Inventor: Hyeong Jin Kim (Incheon, KR)
Assignee: LG ELECTRONICS INC.
H04L51/02G06F3/0481G06F3/04886G06F3/167G06F40/117G06F40/151G06F40/242G06F40/268G06F40/284G06N20/20H04L51/38H04W4/12
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 10,986,046
App. No.
16/555,297
Granted
Apr 20, 2021
Kind
B2
Abstract

Disclosed are a conversation contents summary generating method and a conversation contents summary generating device enabling communication between the conversation contents summary generating device and a user terminal in a 5G communication environment by generating and exposing a summary in a chat room by executing an installed artificial intelligence (AI) algorithm or a machine learning algorithm. A method for generating a summary of conversation contents according to an embodiment of the present disclosure may include collecting conversation texts input by a plurality of conversation participants in a chat room; providing a summary providing interface for outputting a summary of a predetermined range of the conversation texts within the chat room when an operation signal for reviewing a past conversation text is received from the user within a user's chat room including the conversation participants; determining keywords from the predetermined range of the conversation texts within the chat room in response to selection reception by the summary providing interface; generating the summary as a sentence formed by reconstructing the keywords; and exposing the summary on a second surface different from a first surface on which the chat room is exposed, and overlaying and exposing the second surface on the first surface. According to the present disclosure, conversation contents unread and missed by a user in a chat room including a plurality of participants are summarized and provided so that the user may conveniently recognize the conversation contents in the chat room.

Claims (42)

1. A method for generating a summary of conversation contents, the method comprising:

collecting conversation texts input by a plurality of conversation participants in a chat room;

providing a summary providing interface for outputting a summary of a predetermined range of the conversation texts within the chat room when an operation signal for reviewing a past conversation text is received from the user within a user's chat room including the conversation participants;

determining keywords from the predetermined range of the conversation texts within the chat room in response to selection reception by the summary providing interface;

generating the summary as a sentence formed by reconstructing the keywords; and

exposing the summary on a second surface different from a first surface on which the chat room is exposed, and overlaying and exposing the second surface on the first surface,

wherein the generating the summary comprises:

extracting sentences including core keywords from the conversation texts;

calculating a confidence value from a distribution of the core keywords for each of the extracted sentences; and

comparing the confidence value calculated for each of the extracted sentences with a preset threshold value to generate a summary by extracting a sentence having a confidence value equal to or larger than the preset threshold value.

2. The method of claim 1 , wherein the collecting the conversation texts comprises converting and collecting an abbreviation included in the conversation texts into an original word.

3. The method of claim 1 , wherein the collecting the conversation texts comprises converting and collecting an emoticon included in the conversation texts into an emoticon text represented by the emoticon through image analysis of the emoticon.

4. The method of claim 1 , wherein the collecting the conversation texts comprises extracting and collecting a text tagged to an emoticon included in the conversation texts.

5. The method of claim 1 , wherein the providing the summary providing interface comprises providing the summary providing interface for outputting a summary of the conversation texts falling within a range from a section in which a conversation text is not input by the user to a most recent section within the chat room.

6. The method of claim 1 , wherein the providing the summary providing interface comprises providing the summary providing interface for outputting a summary of the conversation texts falling within a range from a conversation text not read immediately before the user enters the chat room to a most recent conversation text.

7. The method of claim 1 , wherein the providing the summary providing interface comprises providing the summary providing interface for outputting the summary of the predetermined range of the conversation texts within the chat room when a touch-related operation signal among scrolling operation signals including a touch and drag operation for reviewing the past conversation text is received from the user within the user's chat room including the conversation participants.

8. The method of claim 1 , wherein the providing the summary providing interface comprises providing an audio summary providing interface for outputting an audio summary of the predetermined range of the conversation texts within the chat room when the operation signal for reviewing the past conversation text is received from the user within the user's chat room including the conversation participants.

9. The method of claim 8 , further comprising generating an audio summary obtained by converting the summary into a sound in response to selection reception by the audio summary providing interface.

10. The method of claim 1 , wherein the determining the keywords comprises:

decomposing the collected conversation texts into morphological units, and extracting at least one noun, at least one interrogative pronoun, and at least one word indicating a tense; and

determining the at least one noun as the keyword, or extracting a word corresponding to an answer to the interrogative pronoun from the collected conversation texts to determine the extracted word as the keyword, or determining the word indicating the tense as the keyword.

11. A device for generating a summary of conversation contents, the device compn sing:

a collecting unit configured to collect conversation texts input by a plurality of conversation participants in a chat room;

a providing unit configured to provide a summary providing interface for outputting a summary of a predetermined range of the conversation texts within the chat room when an operation signal for reviewing a past conversation text is received from the user within a user's chat room including the conversation participants;

a determining unit configured to determine keywords from the predetermined range of the conversation texts within the chat room in response to selection reception by the summary providing interface;

a generating unit configured to generate the summary as a sentence formed by reconstructing the keywords; and

an exposing unit configured to expose the summary on a second surface different from a first surface on which the chat room is exposed, and overlay and expose the second surface on the first surface,

wherein the generating unit is further configured to:

extract sentences including core keywords from the conversation texts;

calculate a confidence value from a distribution of the core keywords for each of the extracted sentences; and

compare the confidence value calculated for each of the extracted sentences with a preset threshold value to generate a summary by extracting a sentence having a confidence value equal to or larger than the preset threshold value.

12. The device of claim 11 , wherein the collecting unit converts and collects an abbreviation included in the conversation texts into an original word.

13. The device of claim 11 , wherein the collecting unit converts and collects an emoticon included in the conversation texts into an emoticon text represented by the emoticon through image analysis of the emoticon.

14. The device of claim 11 , wherein the collecting unit extracts and collects a text tagged to an emoticon included in the conversation texts.

15. The device of claim 11 , wherein the providing unit provides the summary providing interface for outputting a summary of the conversation texts falling within a range from a section in which a conversation text is not input by the user to a most recent section within the chat room.

16. The device of claim 11 , wherein the providing unit provides the summary providing interface for outputting a summary of the conversation texts falling within a range from a conversation text not read immediately before the user enters the chat room to a most recent conversation text.

17. The device of claim 11 , wherein the providing unit provides the summary providing interface for outputting the summary of the predetermined range of the conversation texts within the chat room when a touch-related operation signal among scrolling operation signals including a touch and drag operation for reviewing the past conversation text is received from the user within the user's chat room including the conversation participants.

18. The device of claim 11 , wherein the providing unit provides an audio summary providing interface for outputting an audio summary of the predetermined range of the conversation texts within the chat room when the operation signal for reviewing the past conversation text is received from the user within the user's chat room including the conversation participants.

19. The device of claim 18 , wherein the generating unit generates an audio summary obtained by converting the summary into a sound in response to selection reception by the audio summary providing interface.

20. The device of claim 11 , wherein the determining unit:

decomposes the collected conversation texts into morphological units, and extracts at least one noun, at least one interrogative pronoun, and at least one word indicating a tense; and

determines the at least one noun as the keywords, or extracts a word corresponding to an answer to the interrogative pronoun from the collected conversation texts to determine the extracted word as the keyword, or determines the word indicating the tense as the keyword.

Assignments (3)
RELEASE OF SECURITY INTEREST IN INTELLECTUAL PROPERTY AT REEL/FRAME NO. 55995/0776 Recorded Feb 16, 2022
From: CANADIAN IMPERIAL BANK OF COMMERCE (AS SUCCESSOR IN INTEREST TO WF FUND V LIMITED PARTNERSHIP)
To: TIGERCONNECT, INC. (FORMERLY KNOWN AS TIGERTEXT, INC.)
Reel/Frame 059166/0592 →
SECURITY INTEREST Recorded Apr 21, 2021
From: TIGERCONNECT, INC. (FORMERLY KNOWN AS TIGERTEXT, INC.
To: CANADIAN IMPERIAL BANK OF COMMERCE, AS SUCCESSOR IN INTEREST TO WF FUND V LIMITED PARTNERSHIP
Reel/Frame 055995/0776 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 30, 2019
From: KIM, HYEONG JIN
To: LG ELECTRONICS INC.
Reel/Frame 050226/0729 →
Priority Claims (1)
KR 10-2019-0091847 · Jul 29, 2019 · national
Continuity (1)
Related Publication 20190386937A1 · Dec 19, 2019
Cited By (3)
US 12,230,277 US 12,632,662 US 12,688,222