# US 11,907,172 B2 · Information processing system, information processing method, and recording medium

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

- Number: US 11,907,172 B2
- Document ID: US11907172B2
- Status: Granted
- Kind: B2
- Application number: 17/911,170
- Filed: 2021-03-11
- Published: 2023-04-06
- Granted: 2024-02-20
- Earliest claimed: 2020-03-17
- Assignees: NEC Corporation
- Applicants: NEC Corporation
- Inventors: Junichi Yasuda
- Web page: https://patentgrep.com/patent/US11907172B2

## CPC

- G06F 16/182: Distributed file systems
- G06F 11/2094: {Redundant storage or storage space (G06F11/2056 takes precedence)}
- G06F 11/2097: {maintaining the standby controller/processing unit updated (initialisation or re-synchronisation thereof G06F11/1658 and subgroups)}
- G06F 16/128: {Details of file system snapshots on the file-level, e.g. snapshot creation, administration, deletion (error detection or correction of the data by redundancy in operations or in hardware G06F11/14, G06F11/16)}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

An information processing system preserves data used in machine learning by distributing the data to a plurality of servers, reads setting information indicating a method of partitioning for cross-validation in the machine learning, specifies, based on the setting information, a validation server that executes the cross-validation among the plurality of servers, and validation data which is data used in the cross-validation, specifies an arrangement of the data in the plurality of servers, specifies deficiency data, which is data that is included in the validation data and that is not stored in the validation server, and causes a server that stores the deficiency data among the plurality of servers to transmit the deficiency data to the validation server, based on an arrangement of the specified deficiency data.

## Claims

### Claim 1

An information processing system that preserves data used in machine learning by distributing the data to a plurality of servers, the information processing system comprising:
at least one memory configured to store instructions; and
at least one processor configured to execute the instructions to:
read setting information indicating a method of partitioning for cross-validation in the machine learning;
specify, based on the setting information, a validation server that executes the cross-validation among the plurality of servers, and validation data which is data used in the cross-validation;
specify an arrangement of the data in the plurality of servers;
specify deficiency data, which is data that is included in the validation data and that is not stored in the validation server; and
cause a server that stores the deficiency data among the plurality of servers to transmit the deficiency data to the validation server, based on an arrangement of the specified deficiency data.

### Claim 2 (depends on 1)

The information processing system according to claim 1,
wherein the plurality of servers store master data of the data and a replica of the data in a distributed manner,
the at least one processor is configured to execute the instructions to:
specify an arrangement of the master data and the replica in the plurality of servers; and
specify the deficiency data, which is data included in the validation data and neither the master data nor the replica of which is stored in the validation servers; and
cause the server that stores the deficiency data among the plurality of servers to transmit the replica of the deficiency data to the validation server, based on the arrangement of the specified deficiency data.

### Claim 3 (depends on 1)

The information processing system according to claim 1, further comprising
a learning means that generates a learned model, based on the setting information.

### Claim 4 (depends on 1)

The information processing system according to claim 1,
wherein the setting information indicates the method of partitioning, and the type of data that is a target of the learning or the validation of the machine learning, and
the at least one processor is configured to execute the instructions to specify the validation server, the validation data, and the type, based on the setting information.

### Claim 5 (depends on 1)

The information processing system according to claim 1,
wherein the setting information indicates the method of partitioning, the target data, and a target item that serves as a target of the learning or the validation of the machine learning,
the at least one processor is configured to execute the instructions:
specify the validation server, the validation data, the type, and the target item, based on the setting information; and
specify the deficiency data, which is data that is included in the validation data and that is not stored in the validation server; and
cause the server that stores the deficiency data among the plurality of servers to transmit the target item of the deficiency data to the validation server, based on the arrangement of the specified deficiency data.

### Claim 6

An information processing method executed by an information processing system that preserves data used in machine learning by distributing the data to a plurality of servers, the method comprising:
reading setting information indicating a method of partitioning for cross-validation in the machine learning;
specifying, based on the setting information, a validation server that executes the cross-validation among the plurality of servers, and validation data which is data used in the cross-validation;
specifying an arrangement of the data in the plurality of servers;
specifying deficiency data, which is data that is included in the validation data and that is not stored in the validation server; and
causing a server that stores the deficiency data among the plurality of servers to transmit the deficiency data to the validation server, based on an arrangement of the specified deficiency data.

### Claim 7

A non-transitory recording medium having stored therein a program that causes a computer of an information processing system that preserves data used in machine learning by distributing the data to a plurality of servers, to execute:
reading setting information indicating a method of partitioning for cross-validation in the machine learning;
specifying, based on the setting information, a validation server that executes the cross-validation among the plurality of servers, and validation data which is data used in the cross-validation;
specifying an arrangement of the data in the plurality of servers;
specifying deficiency data, which is data that is included in the validation data and that is not stored in the validation server; and
causing a server that stores the deficiency data among the plurality of servers to transmit the deficiency data to the validation server, based on an arrangement of the specified deficiency data.