# US 2021/0173680 A1 · Artificial intelligence apparatus and method for extracting user's concern

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0173680 A1
- Document ID: US20210173680A1
- Status: Published application
- Kind: A1
- Application number: 16/730,818
- Filed: 2019-12-30
- Published: 2021-06-10
- Earliest claimed: 2019-12-05
- Assignees: LG Electronics Inc.
- Applicants: LG Electronics Inc.
- Inventors: Boosoon JUNG; Soora BANG; Woocheol CHUNG; Bomi JANG
- Web page: https://patentgrep.com/patent/US20210173680A1

## CPC

- G06F 9/453: {Help systems}
- G06F 16/906: Clustering; Classification
- G06F 16/90332: {Natural language query formulation or dialogue systems}
- G06F 16/9535: Search customisation based on user profiles and personalisation
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 20/00: Machine learning
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/045: Combinations of networks
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

The present disclosure provides an artificial intelligence apparatus including a memory configured to store data input by the user or data received by the artificial intelligence apparatus as user interest data, and a processor configured to extract a keyword from the user interest data, classify the keyword as a concern matching the keyword using an interest classification model, and increase an interest rank weight for the concern.

## Claims

### Claim 1

An artificial intelligence apparatus for extracting a user concern, comprising:
a camera configured to capture first image data;
a communication interface configured to transmit and receive data to and from external devices:
a memory configured to store user interest data comprising active data or passive data, wherein the active data corresponds to the first image data input by the user using the camera and the passive data corresponds to second image data received by the artificial intelligence apparatus through the communication interface; and
a processor configured to:
extract a keyword from the user interest data,
classify the extracted keyword as a concern matching the extracted keyword using an interest classification model,
increase an interest rank weight for the classified concern,
determine whether the user interest data corresponds to the active data or the passive data, and
additionally increase the interest rank weight for the classified concern based on a determination that the user interest data corresponds to the active data.

### Claim 2 (depends on 1)

The artificial intelligence apparatus of claim 1, wherein the user interest data comprises at least one of text message data, image data, scheduling data, memo data, search data, or application push notification data.

### Claim 3 (depends on 1)

The artificial intelligence apparatus of claim 1, wherein the keyword is extracted from the user interest data by extracting a noun from text data in the user interest data and extracting the extracted noun as the keyword.

### Claim 4 (depends on 1)

The artificial intelligence apparatus of claim 1, wherein the keyword is extracted from the user interest data based on an object recognized in the first or second image data in the user interest data.

### Claim 5 (depends on 2)

The artificial intelligence apparatus of claim 2, wherein the application push notification data comprises category data and push message data of an application, and
wherein the processor is further configured to extract at least one keyword from the category data and the push message data and to classify the at least one keyword into a concern that matches each of the at least one keyword using the interest classification model.

### Claim 6 (depends on 5)

The artificial intelligence apparatus of claim 5, wherein the processor is further configured to decrease an interest rank weight for a concern matching a keyword extracted from the push notification data of the application when a setting for suspending push notification is input for the application.

### Claim 7

(canceled)

### Claim 8 (depends on 1)

The artificial intelligence apparatus of claim 1, wherein the processor is further configured to tag the user interest data with data about the concern.

### Claim 9 (depends on 8)

The artificial intelligence apparatus of claim 8,
wherein the processor is further configured to group at least one user interest data classified into at least one concern according to concerns and to cause a display of the at least one user interest data according to the at least one concern.

### Claim 10 (depends on 1)

The artificial intelligence apparatus of claim 1, wherein the processor is further configured to determine a recommended service based on the interest rank weight of the concern.

### Claim 11

A method of extracting a user concern, the method comprising:
capturing first image data with a camera;
receiving second image data from external devices through a communication interface;
storing user interest data comprising active data or passive data, wherein the active data corresponds to the first image data input by the user using the camera and the passive data corresponds to second image data received by the artificial intelligence apparatus through the communication interface;
extracting a keyword from the user interest data;
classifying the extracted keyword into a concern matching the extracted keyword using an interest classification model;
increasing an interest rank weight for the classified concern,
determining whether the user interest data corresponds to the active data input by the user using the camera or the passive data received by the artificial intelligence apparatus through the communication interface, and
additionally increasing the interest rank weight for the classified concern based on a determination that the user interest data corresponds to the active data.

### Claim 12 (depends on 11)

The method of claim 11, wherein the user interest data comprises at least one of text message data, image data, scheduling data, memo data, search data, or application push notification data.

### Claim 13 (depends on 11)

The method of claim 11, wherein the keyword is extracted by extracting a noun from text data included in the user interest data and extracting the extracted noun as the keyword.

### Claim 14 (depends on 11)

The method of claim 11, wherein the keyword is extracted from the user interest data based on an object recognized in the first or second image data included in the user interest data.

### Claim 15 (depends on 12)

The method of claim 12, wherein the application push notification data comprises at least one of name data, category data, or push message data of an application, and
wherein the extracting of the keyword includes extracting at least one keyword from the category data and the push message data included in the application push notification data.

### Claim 16 (depends on 15)

The method of claim 15, further comprising:
receiving a setting for suspending reception of a push notification for the application; and
decreasing an interest rank weight for a concern matching a keyword extracted from the push notification data of the application,

### Claim 17

(canceled)

### Claim 18 (depends on 11)

The method of claim 11, further comprising:
tagging the user interest data with data about the concern.

### Claim 19 (depends on 18)

The method of claim 18, further comprising:
grouping at least one user interest data classified into at least one concern according to concerns and displaying the at least one user interest data according to the at least one concern.

### Claim 20 (depends on 11)

The method of claim 11, further comprising:
determining a recommended service based on the interest rank weight of the concern.
