Response determining device and controlling method of electronic device therefor

Patent metadata

Field	Value
Publication number	US-12277149-B2
Application number	US-202217712301-A
Country	US
Kind code	B2
Filing date	Apr 4, 2022
Priority date	Jan 22, 2021
Publication date	Apr 15, 2025
Grant date	Apr 15, 2025

Abstract

A method of controlling an electronic device includes, based on obtaining first text information corresponding to a user query, identifying a main query corresponding to the first text information; obtaining a plurality of responses related to the main query; identifying a plurality of phrases included in the first text information; identifying at least one first phrase corresponding to the main query among the plurality of phrases based on a similarity between the plurality of phrases and the main query; identifying, among the plurality of responses, at least one first response corresponding to a remaining second phrase, among the plurality of phrases except the at least one first phrase, based on a similarity between each of the plurality of responses and the remaining second phrase; and providing second text information corresponding to a remaining second response other than the at least one first response among the plurality of responses.

First claim

What is claimed is:

1. An electronic device for providing artificial intelligence (AI), guidance for resolving symptoms of an electronic product, comprising:
a display;
one or more processors; and
memory storing instructions that, when executed by the one or more processors, cause the electronic device to:
display a user interface (UI) on the display for obtaining a user input indicating a user query comprising a plurality of sentences, wherein the plurality of sentences comprise at least one symptom of the electronic product;
determine first text information corresponding to the user query;
identify, based on determining the first text information, a main query corresponding to the first text information, wherein the main query is determined based on at least one of a database comprising a plurality of predefined main queries or a first neural network model trained to identify one or more main queries corresponding to the first text information, and wherein the main query corresponds to a first symptom from among the at least one symptom;
obtain a plurality of responses related to the main query from the database, wherein the database further comprises a first plurality of responses corresponding to the plurality of predefined main queries, and wherein the first plurality of responses comprises at least one from among:
a response query corresponding second symptom of the electronic product, and
a response corresponding to self-action to resolve the first symptom;
identify a plurality of phrases included in the first text information based on at least one of a conjunction, an ending mark, a punctuation mark, or based on a second neural network model trained with a labeled data set of text information;
identify at least one first phrase corresponding to the main query among the plurality of phrases based on a first similarity in a first vector space between each of the plurality of phrases and the main query, wherein the first vector space is constructed based on a feature extraction model or word embedding model, the first vector space comprises the plurality of phrases and the main query, and the first similarity is determined based on positions in the first vector space of the plurality of phrases compared to the main query;
identify, among the plurality of responses, at least one first response corresponding to a remaining second phrase, among the plurality of phrases except the at least one first phrase, based on a second similarity in a second vector space between each of the plurality of responses and the remaining second phrase, wherein the second vector space comprises the plurality of responses and the remaining second phrase, and the second similarity is determined based on positions in the second vector space of the plurality of responses compared to the remaining second phrase, and wherein the at least one first response comprises at least one response query corresponds to at least one third symptom of the electronic product;
identify at least one context corresponding to the remaining second phrase among a plurality of predefined contexts;
identify the at least one first response, among the plurality of responses, corresponding to the at least one context;
filter the plurality of response to exclude the at least one first response; and
display, in the user interface, second text information corresponding to a remaining second response, the remaining second response corresponding to at least one fourth symptom of the electronic product and being a response other than the at least one first response among the plurality of responses, wherein the remaining second response is determined based on positions in the second vector space of the plurality of responses compared to the remaining second response;

wherein the electronic device further comprises a microphone and a speaker, and
wherein the one or more processors are configured to execute the instructions to cause the electronic device to:
receive a first voice signal corresponding to the user query through the microphone,
obtain the first text information based on the first voice signal,
obtain a second voice signal corresponding to the at least one first response based on text information corresponding to the at least one first response, and
output the second voice signal through the speaker.

2. The electronic device of claim 1 , wherein the one or more processors are configured to execute the instructions to cause the electronic device to:
obtain, with respect to each of the plurality of phrases, a first score based on the first similarity between the main query and each of the plurality of phrases, and
identify the at least one first phrase, among the plurality of phrases, of which the first score is greater than or equal to a preset first threshold value.

3. The electronic device of claim 1 , wherein the one or more processors are configured to execute the instructions to cause the electronic device to:
obtain, with respect to each of the plurality of responses, a second score based on the second similarity between the remaining second phrase and each of the plurality of responses, and
identify the at least one first response, among the plurality of responses, of which the second score is greater than or equal to a preset second threshold value.

4. The electronic device of claim 1 , wherein the one or more processors are configured to execute the instructions to cause the electronic device to identify the plurality of phrases included in the first text information based on at least one of a conjunction, an ending mark, or a punctuation included in the first text information.

5. The electronic device of claim 1 , wherein the one or more processors are configured to execute the instructions to cause the electronic device to, based on information indicative of positivity or negativity for the at least one first response being included in the remaining second phrase, provide the at least one first response and the information indicative of positivity or negativity for the at least one first response along with the second text information corresponding to the remaining second response.

6. The electronic device of claim 1 , further comprising:
a communicator interface,
wherein the one or more processors are configured to execute the instructions to cause the electronic device to:
receive the first text information from an external device through the communicator interface, and
control the communicator interface to transmit the second text information corresponding to the remaining second response to the external device.

7. The electronic device of claim 1 ,
wherein the one or more processors are configured to execute the instructions to cause the electronic device to:
receive a user input related to the remaining second response through the UI.

8. A method of controlling an electronic device, the method comprising:
displaying a user interface (UI) on a display for obtaining a user input indicating a user query comprising a plurality of sentences, wherein the plurality of sentences comprise at least one symptom of the electronic product;
determining first text information corresponding to the user query;
identifying, based on determining the first text information, a main query corresponding to the first text information, wherein the main query is determined based on at least one of a database comprising a plurality of predefined main queries or a first neural network model trained to identify one or more main queries corresponding to the first text information, and wherein the main query corresponds to a first symptom from among the at least one symptom;
obtaining a plurality of responses related to the main query from the database, wherein the database f

Assignees

• Samsung Electronics Co Ltd

Inventors

• An Jaehyung
• Kim Hoyoung
• Yang Dongil
• Lee Jiyeon
• Jung Cheolseung

Classifications

• G06F16/3344

  using natural language analysis · CPC title

• G06F40/30 Primary

  Semantic analysis · CPC title

• G06F40/289

  Phrasal analysis, e.g. finite state techniques or chunking · CPC title

• Y10S715/978

  Audio interaction as part of an operator interface · CPC title

• G06F40/211

  Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars · CPC title