# US 10,768,868 B2 · System and method for performing fault diagnosis on a device

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

- Number: US 10,768,868 B2
- Document ID: US10768868B2
- Status: Granted
- Kind: B2
- Application number: 16/460,495
- Filed: 2019-07-02
- Published: 2020-01-09
- Granted: 2020-09-08
- Earliest claimed: 2018-07-09
- Assignees: Canon Kabushiki Kaisha
- Applicants: Canon Kabushiki Kaisha
- Inventors: Shunsuke Ota
- Web page: https://patentgrep.com/patent/US10768868B2

## CPC

- G06F 3/121: {Facilitating exception or error detection and recovery, e.g. fault, media or consumables depleted}
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06F 3/1234: {Errors handling and recovery, e.g. reprinting (G06F3/1261 takes precedence)}
- G06F 3/1285: {Remote printer device, e.g. being remote from client or server}
- G06F 11/0709: {in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems}
- G06F 11/0736: {in functional embedded systems, i.e. in a data processing system designed as a combination of hardware and software dedicated to performing a certain function (testing or monitoring of automated control systems G05B23/02)}
- G06F 11/0751: {Error or fault detection not based on redundancy (power supply failures G06F1/30; network fault management H04L41/06)}
- G06F 11/0787: {Storage of error reports, e.g. persistent data storage, storage using memory protection}
- G06F 11/079: {Root cause analysis, i.e. error or fault diagnosis (in a hardware test environment G06F11/22; in a software test environment G06F11/36)}
- G06N 20/00: Machine learning
- H04L 41/065: {involving logical or physical relationship, e.g. grouping and hierarchies}
- H04L 41/0686: Additional information in the notification, e.g. enhancement of specific meta-data
- H04L 41/142: using statistical or mathematical methods
- H04L 41/16: using machine learning or artificial intelligence
- G06F 3/1288: {in client-server-printer device configuration}
- G06F 11/3447: {Performance evaluation by modeling}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

In connection with an error event detected in an image forming apparatus, a fault diagnosis service server diagnoses the error event and obtains a diagnosis result for handling to be performed on the error event detected in the image forming apparatus. In a case where the diagnosis result indicates a plurality of candidates for handling to be performed on the error event, the fault diagnosis service server compares the plurality of candidates for handling to a result estimated by a diagnosis model with respect to the error event. In a case where the plurality of candidates for handling is different from the result estimated by the diagnosis model, the fault diagnosis service obtains partial operation information from a history of the image forming apparatus and executes machine relearning based on the obtained partial operation information with respect to a layer in a latter stage of a plurality of layers included in the diagnosis model, each layer performing a different estimation.

## Claims

### Claim 1

A system comprising:
a storage storing a model for estimating an appropriate handling for an error event detected in a network device, the model constructed by machine learning using operation information collected from the network device and a history of handlings for the network device;
at least one memory storing instructions; and
at least one processor which, when executing the instructions, causes the system to:
obtain, based on the error event detected in the network device, a diagnosis result of diagnosis in the network device for identifying handling to be performed on the error event;
in a case where the diagnosis result indicates a plurality of candidates for handling to be performed on the error event, compare each of the plurality of candidates for handling the error event, to a result estimated with respect to the error event using the model; and
in a case where each of the plurality of candidates for handling the error event is found by the comparing to be different from the estimated result, obtain partial operation information from the history; and
execute machine relearning based on the obtained partial operation information with respect to a layer in a latter stage in a plurality of layers included in the model, each layer performing a different estimation.

### Claim 2 (depends on 1)

The system according to claim 1,
wherein, in the history, the handling for the network device is associated with the operation information, and
wherein the partial operation information to be used for the machine relearning is operation information associated with, in the history, a same handling for the network device as one of the plurality of candidates.

### Claim 3 (depends on 1)

The system according to claim 1, wherein the machine relearning is executed with respect to a layer in a final stage in the plurality of layers.

### Claim 4 (depends on 3)

The system according to claim 3, wherein a node of the layer in the final stage is associated with the plurality of candidates for handling, and a weight from a previous layer to the node of the layer in the final stage is adjusted in the machine relearning.

### Claim 5 (depends on 1)

The system according to claim 1, wherein the handling represents a specific measure for a specific part in the network device or a specific operation for the network device.

### Claim 6 (depends on 1)

The system according to claim 1, wherein a result having a highest probability or a result having a probability within a predetermined threshold value from the highest probability is regarded as an estimation result in estimation using the model.

### Claim 7

A method performed in a system storing a model for estimating an appropriate handling for an error event detected in a network device, the appropriate handling estimated based on operation information collected from the network device and a history of handlings for the network device, the method comprising:
obtaining, based on an error event detected in the network device, a diagnosis result of diagnosis in the network device for identifying handling to be performed on the error event;
in a case where the diagnosis result indicates a plurality of candidates for handling to be performed on the error event, comparing each of the plurality of candidates for handling, to a result estimated with respect to the error event using the model;
in a case where each of the plurality of candidates for handling the error event is found by the comparing to be different from the estimated result, obtaining partial operation information from the history; and
executing machine relearning based on the obtained partial operation information with respect to a layer in a latter stage in a plurality of layers included in the model, each performing a different estimation.

### Claim 8

A non-transitory computer readable storage medium on which is stored a computer program for causing a computer to execute a method for a system that stores a model for estimating an appropriate handling for an error event detected in a network device, the model constructed by machine learning using operation information collected from the network device and a history of handlings for the network device, the method comprising:
obtaining, based on an error event detected in the network device, a diagnosis result of diagnosis in the network device for identifying handling to be performed on the error event;
in a case where the diagnosis result indicates a plurality of candidates for handling to be performed on the error event, comparing each of the plurality of candidates for handling, to a result estimated with respect to the error event using the model;
in a case where each of the plurality of candidates for handling the error event is found by the comparing to be different from the estimated result, obtaining partial operation information from the history; and
executing machine relearning based on the obtained partial operation information with respect to a layer in a latter stage in a plurality of layers included in the model, each performing a different estimation.