# US 12,602,590 B2 · Artificial intelligence system that collects interactive picture data for building artificial intelligence data and analyzes correlation between art competency and academic achievement

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

- Number: US 12,602,590 B2
- Document ID: US12602590B2
- Status: Granted
- Kind: B2
- Application number: 18/350,956
- Filed: 2023-07-12
- Published: 2023-11-02
- Granted: 2026-04-14
- Earliest claimed: 2021-09-03
- Assignees: I-Scream Arts Co., Ltd.
- Applicants: I-Scream Arts Co., Ltd.
- Inventors: Ji-Hoon Kim; Seung-Ah Kim
- Web page: https://patentgrep.com/patent/US12602590B2

## CPC

- G06N 3/091: Active learning
- A61B 5/00: Measuring for diagnostic purposes (radiation diagnosis A61B6/00; diagnosis by ultrasonic, sonic or infrasonic waves A61B8/00); Identification of persons
- A61B 5/16: Devices for psychotechnics (using teaching or educational appliances G09B1/00 - G09B7/00); Testing reaction times {; Devices for evaluating the psychological state}
- G06N 3/045: Combinations of networks
- G06N 3/08: Learning methods
- G06V 10/24: Aligning, centring, orientation detection or correction of the image
- G16H 10/20: for electronic clinical trials or questionnaires
- G16H 20/70: relating to mental therapies, e.g. psychological therapy or autogenous training
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

The present disclosure relates to a technique that collects picture data created by a user and predicts an academic achievement of the user through the picture data and an artificial intelligence system that collects interactive picture data for building artificial intelligence data and analyzes correlation between an art competency and an academic achievement according to an exemplary embodiment of the present disclosure includes a picture data collector which provides an interactive environment to a user to induce the user to perform a process of labeling an object included in picture data created by the user and collects picture data including an object whose class is annotated by the labeling process; an academic achievement prediction unit which evaluates an art competency of the user using the picture data and predicts an academic achievement of the user through a learning competency with a high correlation with an art competency of the user; and a database in which picture data to be used to evaluate an art competency of the user by the academic achievement prediction unit is stored.

## Claims

### Claim 1

An artificial intelligence system that collects interactive picture data for building artificial intelligence data and analyzes correlation between an art competency and an academic achievement, comprising:
a picture data collector which provides an interactive environment to a user to induce the user to perform a process of labeling an object included in picture data created by the user and collects the picture data including the object whose class is annotated by the labeling process;
an academic achievement prediction processor which evaluates the art competency of the user using the picture data and predicts the academic achievement of the user through a learning competency with a correlation with the art competency of the user; and
a database in which the picture data to be used to evaluate the art competency of the user by the academic achievement prediction processor is stored,
wherein the picture data collector includes:
a user interface for providing the interactive environment to the user;
an object extractor which is trained with a first artificial intelligence model to extract the object from the picture data created by the user through the user interface;
a text converter which converts a voice recorded to explain the object by the user into a text after generating the object in the picture data; and
an object labeling processor which is trained with a second artificial intelligence model which extracts a sentence or a phrase from the text to summarize the text, annotates the class of the object by labeling the object with a key word which forms the sentence or the phrase, and transmits the picture data of the user in which the class of the object is annotated to the database.

### Claim 2 (depends on 1)

The artificial intelligence system of claim 1, wherein the first artificial intelligence model sets and extracts a boundary area in a position where the object is disposed, from the picture data to generate a quadrangular coordinate for the position of the object and modifies an image in the boundary area in accordance with a size of a CNN input layer and then extracts a feature vector and annotates the class of the object through the extraction of the feature vector.

### Claim 3 (depends on 1)

The artificial intelligence system of claim 1, wherein during the converting of the voice into the text in the text converter, if each word included in the sentence or the phrase that forms the text is a standard language, the object labeling processor annotates the class of the object based on extractive summarization of the second artificial intelligence model.

### Claim 4 (depends on 1)

The artificial intelligence system of claim 1, wherein the object labeling processor is trained by a third artificial intelligence model to abstractively summarize the text by generating a new sentence which is not present in the text based on a key context of the text and annotate the class of the object by labeling the object with the key word which forms the sentence.

### Claim 5 (depends on 4)

The artificial intelligence system of claim 4, wherein during the converting of the voice into the text in the text converter, if one or more words which are not standard language are included in the sentence or the phrase that forms the text, the object labeling processor annotates the class of the object based on abstractive summarization of the third artificial intelligence model.

### Claim 6 (depends on 1)

The artificial intelligence system of claim 1, wherein the text converter receives a recording file in which the voice is recorded, through the user interface and preprocesses the recording file by removing a noise recorded together in the recording file during the process of recording the voice, extracts the voice of the recording file as the text, and then modifies the words of the text in accordance with a standard language.

### Claim 7 (depends on 1)

The artificial intelligence system of claim 1, wherein the academic achievement prediction processor includes:
a search processor which searches for the picture data of the user in which the class of the object transmitted from the object labeling processor to the database is annotated from the database;
an evaluation processor which evaluates the art competency of the user from the picture data of the user in which the class of the object searched by the search processor is annotated; and
a prediction processor which when information about the learning competency with the correlation with the art competency of the user evaluated by the evaluation processor is searched by the search processor, receives the information about the learning competency from the search processor, and predicts the academic achievement of the user based on the information about the learning competency.

### Claim 8 (depends on 7)

The artificial intelligence system of claim 7, wherein when the search processor searches for the picture data of the user in which the class of the object is annotated from the database and receives the information about the art competency of the user from the evaluation processor, the search processor is trained with a fourth artificial intelligence model to search the learning competency with the correlation with the art competency of the user from the database.

### Claim 9 (depends on 8)

The artificial intelligence system of claim 8, wherein the search processor is trained with the fourth artificial intelligence model to search for actual photograph data of the object to be compared, for the evaluation processor to evaluate the art competency of the user, from the database to transmit the data to the evaluation processor.

### Claim 10 (depends on 9)

The artificial intelligence system of claim 9, wherein the evaluation processor is trained with a fifth artificial intelligence model to compare the picture data of the user in which the class of the object is annotated and the actual photograph data of the object, and rate a score to the picture data of the user in which the class of the object is annotated, based on a degree of describing the object of the actual photograph data by the object of the picture data, and evaluate the art competency of the user based on the score.

### Claim 11 (depends on 10)

The artificial intelligence system of claim 10, wherein the prediction processor is trained with a sixth artificial intelligence model to predict the academic achievement of the user based on the information about the learning competency when the learning competency with the correlation with the art competency of the user calculated based on the score is received from the search processor.