# US 2026/0037557 A1 · Non-transitory computer-readable recording medium having stored therein prediction program, information processing apparatus, and computer-implemented prediction method

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

- Number: US 2026/0037557 A1
- Document ID: US20260037557A1
- Status: Published application
- Kind: A1
- Application number: 19/254,963
- Filed: 2025-06-30
- Published: 2026-02-05
- Earliest claimed: 2024-07-31
- Assignees: Fujitsu Limited
- Applicants: Fujitsu Limited
- Inventors: Satoru KODA
- Web page: https://patentgrep.com/patent/US20260037557A1

## CPC

- G06F 16/3334: {Selection or weighting of terms from queries, including natural language queries}
- G06F 40/284: Lexical analysis, e.g. tokenisation or collocates
- G06F 16/334: Query execution (filtering based on additional data G06F16/335)
- G06F 40/279: Recognition of textual entities
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A non-transitory computer-readable recording medium having stored therein a prediction program that causes a computer to execute a process including allocating an input first character string to a block that satisfies a predetermined condition, predicting, by using a feature amount of each character of a second character string in each block and a detector configured to detect keyword stuffing, a probability that keyword stuffing is present in the second character string, and predicting a center and a length of a keyword segment in the second character string when the probability is a predetermined threshold or more.

## Claims

### Claim 1

A non-transitory computer-readable recording medium having stored therein a prediction program that causes a computer to execute a process comprising:
allocating an input first character string to a block that satisfies a predetermined condition;
predicting, by using a feature amount of each character of a second character string in each block and a detector configured to detect keyword stuffing, a probability that keyword stuffing is present in the second character string; and
predicting a center and a length of a keyword segment in the second character string when the probability is a predetermined threshold or more.

### Claim 2 (depends on 1)

The non-transitory computer-readable recording medium according to claim 1,
wherein the predicting of the probability is to predict a probability that a true center of the keyword stuffing is present in the second character string.

### Claim 3 (depends on 1)

The non-transitory computer-readable recording medium according to claim 1, further comprising:
correcting a boundary position indicating at least one of a start position and an end position in the first character string that is obtained based on the predicted center and the predicted length of the keyword segment to an adjacent position indicating a position of a character positioned adjacent to the character of the boundary position with a corrector configured to correct the boundary position.

### Claim 4 (depends on 2)

The non-transitory computer-readable recording medium according to claim 2, further comprising:
correcting a boundary position indicating at least one of a start position and an end position in the first character string that is obtained based on the predicted center and the predicted length of the keyword segment to an adjacent position indicating a position of a character positioned adjacent to the character of the boundary position with a corrector configured to correct the boundary position.

### Claim 5 (depends on 1)

The non-transitory computer-readable recording medium according to claim 1,
wherein the allocating to the block is to allocate at least a part of the second character string in each adjacent block in a manner of overlapping each other.

### Claim 6 (depends on 2)

The non-transitory computer-readable recording medium according to claim 2,
wherein the allocating to the block is to allocate at least a part of the second character string in each adjacent block in a manner of overlapping each other.

### Claim 7 (depends on 2)

The non-transitory computer-readable recording medium according to claim 2, further comprising:
training the detector so that the center of the keyword segment matches the true center of the keyword stuffing.

### Claim 8

An information processing apparatus with a processor that execute a process comprising:
allocating an input first character string to a block that satisfies a predetermined condition;
predicting, by using a feature amount of each character of a second character string in each block and a detector configured to detect keyword stuffing, a probability that keyword stuffing is present in the second character string; and
predicting a center and a length of a keyword segment in the second character string when the probability is a predetermined threshold or more.

### Claim 9 (depends on 8)

The information processing apparatus according to claim 8,
wherein a process of predicting the probability is to predict a probability that a true center of the keyword stuffing is present in the second character string.

### Claim 10 (depends on 8)

The information processing apparatus according to claim 8,
wherein the processor corrects a boundary position indicating at least one of a start position and an end position in the first character string that is obtained based on the predicted center and the predicted length of the keyword segment to an adjacent position indicating a position of a character positioned adjacent to the character of the boundary position with a corrector configured to correct the boundary position.

### Claim 11 (depends on 9)

The information processing apparatus according to claim 9,
wherein the processor corrects a boundary position indicating at least one of a start position and an end position in the first character string that is obtained based on the predicted center and the predicted length of the keyword segment to an adjacent position indicating a position of a character positioned adjacent to the character of the boundary position with a corrector configured to correct the boundary position.

### Claim 12 (depends on 8)

The information processing apparatus according to claim 8,
wherein the allocating to the block is to allocate at least a part of the second character string in each adjacent block in a manner of overlapping each other.

### Claim 13 (depends on 9)

The information processing apparatus according to claim 9,
wherein the allocating to the block is to allocate at least a part of the second character string in each adjacent block in a manner of overlapping each other.

### Claim 14 (depends on 9)

The information processing apparatus according to claim 9,
wherein the processor trains the detector so that the center of the keyword segment matches the true center of the keyword stuffing.

### Claim 15

A computer-implemented prediction method that causes a computer to execute a process comprising:
allocating an input first character string to a block that satisfies a predetermined condition;
predicting, by using a feature amount of each character of a second character string in each block and a detector configured to detect keyword stuffing, a probability that keyword stuffing is present in the second character string; and
predicting a center and a length of a keyword segment in the second character string when the probability is a predetermined threshold or more.

### Claim 16 (depends on 15)

The computer-implemented prediction method according to claim 15,
wherein the predicting of the probability is to predict a probability that a true center of the keyword stuffing is present in the second character string.

### Claim 17 (depends on 15)

The computer-implemented prediction method according to claim 15, further comprising:
correcting a boundary position indicating at least one of a start position and an end position in the first character string that is obtained based on the predicted center and the predicted length of the keyword segment to an adjacent position indicating a position of a character positioned adjacent to the character of the boundary position with a corrector configured to correct the boundary position.

### Claim 18 (depends on 16)

The computer-implemented prediction method according to claim 16, further comprising:
correcting a boundary position indicating at least one of a start position and an end position in the first character string that is obtained based on the predicted center and the predicted length of the keyword segment to an adjacent position indicating a position of a character positioned adjacent to the character of the boundary position with a corrector configured to correct the boundary position.

### Claim 19 (depends on 15)

The computer-implemented prediction method according to claim 15,
wherein the allocating to the block is to allocate at least a part of the second character string in each adjacent block in a manner of overlapping each other.

### Claim 20 (depends on 16)

The computer-implemented prediction method according to claim 16, further comprising:
training the detector so that the center of the keyword segment matches the true center of the keyword stuffing.
