# US 12,770,527 B2 · Method, system, and non-transitory computer-readable recording medium for estimating concept understanding

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

- Number: US 12,770,527 B2
- Document ID: US12770527B2
- Status: Granted
- Kind: B2
- Application number: 18/096,831
- Filed: 2023-01-13
- Published: 2023-07-27
- Granted: 2026-10-20
- Earliest claimed: 2022-01-13
- Assignees: Mata Edu, Inc.
- Applicants: Mata Edu, Inc.
- Inventors: Tae Hyoung Oh; Myunghoon Ahn; Du Seop Jung; Sang Hoon Lee; Jaehyeong Ahn
- Web page: https://patentgrep.com/patent/US12770527B2

## CPC

- G09B 7/00: Electrically-operated teaching apparatus or devices working with questions and answers (mechanically operated G09B3/00; computing arrangements G06F)
- G06Q 50/205: {Education administration or guidance}
- G09B: EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS (devices for psychotechnics or for testing reaction times A61B5/16; games, sports, amusements A63; projectors, projector screens G03B)
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

A method for estimating concept understanding is provided. The method includes the steps of: generating concept-specific correctness/incorrectness sequence data with respect to at least one user, with reference to data on a result of the at least one user solving at least one question associated with at least one concept; and estimating a first user's understanding of a first concept using a concept-specific understanding estimation model that is trained on the basis of the concept-specific correctness/incorrectness sequence data.

## Claims

### Claim 1

A method performed in a system for estimating concept understanding, the system comprising one or more processors and the method comprising the steps of:
by the one or more processors, performing concept-specific categorization for at least one question associated with at least one concept with reference to a lookup table using a natural language processing (NLP) algorithm and a clustering algorithm;
by the one or more processors, generating concept-specific correctness/incorrectness sequence data represented in a matrix structure with respect to at least one user, with reference to data on a result of the at least one user solving the at least one question and the concept-specific categorization for the at least one question;
by the one or more processors, training a concept-specific understanding estimation model on the basis of the concept-specific correctness/incorrectness sequence data in the matrix structure; and
by the one or more processors, estimating a first user's understanding of a first concept using the trained concept-specific understanding estimation model,
wherein the concept-specific correctness/incorrectness sequence data includes first sequence data generated at a first time point and second sequence data generated at a second time point that follows the first time point by a predetermined amount of time,
wherein the concept-specific understanding estimation model is trained such that the concept-specific understanding is estimated by assigning a greater weight to the second sequence data generated at the second time point than to the first sequence data generated at the first time point,
wherein a weight assigned to lth sequence data out of a total of t pieces of sequence data is determined on the basis of a value obtained by dividing a t−lth power of a specific constant by a geometric series from a first term to a (t−1)th term for the t−lth power of the specific constant, and
wherein in the estimating step, similarity between a concept not encountered by the first user and a concept encountered by the first user is assessed, and the first user's understanding of the concept not encountered by the first user is estimated from the first user's understanding of the concept encountered by the first user on the basis of the similarity.

### Claim 2 (depends on 1)

The method of claim 1, wherein the concept-specific understanding estimation model is trained using a Bayesian knowledge tracing algorithm.

### Claim 3 (depends on 1)

The method of claim 1, further comprising the step of:
by the one or more processors, assessing the concept-specific understanding estimation model using a result of the estimation.

### Claim 4 (depends on 1)

A non-transitory computer-readable recording medium having stored thereon a computer program for executing the method of claim 1.

### Claim 5 (depends on 1)

The method of claim 1, wherein in the estimating step, the first user's understanding of the concept not encountered by the first user is further estimated using a matrix factorization algorithm on the concept-specific correctness/incorrectness sequence data represented in the matrix structure with respect to a plurality of concepts including the concept encountered by the first user and the concept not encountered by the first user.

### Claim 6

A system for estimating concept understanding, the system comprising one or more processors configured to:
perform concept-specific categorization for at least one question associated with at least one concept with reference to a lookup table using a natural language processing (NLP) algorithm and a clustering algorithm;
generate concept-specific correctness/incorrectness sequence data represented in a matrix structure with respect to at least one user, with reference to data on a result of the at least one user solving the at least one question and the concept-specific categorization for the at least one question;
train a concept-specific understanding estimation model on the basis of the concept-specific correctness/incorrectness sequence data in the matrix structure; and
estimate a first user's understanding of a first concept using the trained concept-specific understanding estimation model,
wherein the concept-specific correctness/incorrectness sequence data includes first sequence data generated at a first time point and second sequence data generated at a second time point that follows the first time point by a predetermined amount of time, and
wherein the concept-specific understanding estimation model is trained such that the concept-specific understanding is estimated by assigning a greater weight to the second sequence data generated at the second time point than to the first sequence data generated at the first time point,
wherein a weight assigned to lth sequence data out of a total of t pieces of sequence data is determined on the basis of a value obtained by dividing a t−lth power of a specific constant by a geometric series from a first term to a (t−1)th term for the t−lth power of the specific constant, and
wherein the one or more processors are configured to assess similarity between a concept not encountered by the first user and a concept encountered by the first user, and estimate the first user's understanding of the concept not encountered by the first user from the first user's understanding of the concept encountered by the first user on the basis of the similarity.

### Claim 7 (depends on 6)

The system of claim 6, wherein the concept-specific understanding estimation model is trained using a Bayesian knowledge tracing algorithm.

### Claim 8 (depends on 6)

The system of claim 6, wherein the one or more processors are configured to assess the concept-specific understanding estimation model using a result of the estimation.

### Claim 9 (depends on 6)

The system of claim 6, wherein the one or more processors are configured to further estimate the first user's understanding of the concept not encountered by the first user using a matrix factorization algorithm on the concept-specific correctness/incorrectness sequence data represented in the matrix structure with respect to a plurality of concepts including the concept encountered by the first user and the concept not encountered by the first user.
