# US 2022/0019496 A1 · Error documentation assistance

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

- Number: US 2022/0019496 A1
- Document ID: US20220019496A1
- Status: Published application
- Kind: A1
- Application number: 17/370,968
- Filed: 2021-07-08
- Published: 2022-01-20
- Earliest claimed: 2020-07-14
- Assignees: State Farm Mutual Automobile Insurance Company
- Applicants: State Farm Mutual Automobile Insurance Company
- Inventors: Javier Lozano; William Luo; Carlos Gonzalez; Bing (Aaron) Lin
- Web page: https://patentgrep.com/patent/US20220019496A1

## CPC

- 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 20/20: Ensemble learning
- G06F 11/0772: {Means for error signaling, e.g. using interrupts, exception flags, dedicated error registers}
- G06F 11/0778: {Dumping, i.e. gathering error/state information after a fault for later diagnosis}
- G06F 11/0781: {Error filtering or prioritizing based on a policy defined by the user or on a policy defined by a hardware/software module, e.g. according to a severity level}
- G06F 11/0793: {Remedial or corrective actions (recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40)}
- G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06N 20/00: Machine learning
- G06N 5/022: Knowledge engineering; Knowledge acquisition
- G06N 5/04: Inference or reasoning models
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

An error documentation system including tools to collect and analyze application error data for individual development teams and tools to share documented defects and solutions across development teams during any stage of development cycle. The system may receive and analyze event logs for error events triggered by applications on end-user devices. The system may automatically generate defect tickets and/or ticket entries for defects identified in event logs. The system may train one or more machine learning (ML) models to correlate input with identified defects from a defects database. In response to identifying correlated identified defects, the system may generate ticket entries indicating the correlated identified defects and associated solutions for the defects. The system may provide an interface for users to query the data stored in the database.

## Claims

### Claim 1

A system comprising:
one or more processors; and
computer-readable media storing computer-executable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving, from one or more computing devices, an event log associated with an error event, the event log including data logged in response to the error event triggered on an application;
identifying, based at least in part on the event log, a defect and corresponding defect information;
determining, by inputting the defect information into one or more correlation models, the defect correlates to a resolved defect identified in a defects database;
identifying a solution of the resolved defect indicated in the defects database;
generating a defect ticket for the defect including the defect information and indicating the solution; and
storing the defect ticket in the defects database.

### Claim 2 (depends on 1)

The system of claim 1, the defect information including one or more of: an error type, an error message, a sequence log, a response time of a request, a sequence code, a stack trace, an exposed endpoint, an application identifier, a stage of development cycle, and a severity level.

### Claim 3 (depends on 1)

The system of claim 1, the operations further comprising:
generating a task to request review of the defect ticket;
generating a notification for the task;
publishing the notification to a subscriber of events associated with the application; and
sending the notification to a device associated with the subscriber.

### Claim 4 (depends on 3)

The system of claim 3, wherein determining the defect correlates to the resolved defect includes:
generating a confidence score associated with the defect correlating to the resolved defect; and
determining the confidence score is above a threshold.

### Claim 5 (depends on 4)

The system of claim 4, the operations further comprising:
receiving confirmation that the solution is a resolution for the defect;
creating training data that includes the defect ticket, the solution, and the confirmation; and
retraining the one or more models using training data.

### Claim 6 (depends on 1)

The system of claim 1, the operations further comprising:
receiving a query indicating one of an error type or an error message; and
retrieving, from the defects database, one or more solutions associated with the query.

### Claim 7 (depends on 1)

The system of claim 1, the operations further comprising:
receiving, from the one or more computing devices, an additional event log associated with an additional error event;
determining an additional defect associated with the additional event log matches the defect; and
adding a log identifier associated with the additional event log to the defect ticket.

### Claim 8 (depends on 7)

The system of claim 7, the operations further comprising:
determining a count of log identifiers associated with the defect ticket exceeds a threshold; and
increasing a priority level of the defect ticket based at least in part on the count of log identifiers exceeding the threshold.

### Claim 9

A method, comprising:
training a correlation model with training data to correlate input data to identified defects and to output associated confidence scores;
receiving an event log associated with an error event, the event log including data logged in response to the error event, and the error event being detected on an application;
identifying, based at least in part on the event log, a defect and corresponding defect information;
determining, by inputting the defect information into the correlation model, the defect correlates to an identified defect from a defects database;
generating a confidence score associated with the defect correlating to the identified defect;
determining the confidence score is above a threshold; and
generating a defect ticket for the defect and indicating the identified defect.

### Claim 10 (depends on 9)

The method of claim 9, further comprising:
identifying a solution of the identified defect indicated in the defects database; and
indicating the solution on the defect ticket.

### Claim 11 (depends on 10)

The method of claim 10, further comprising:
generating a task to request review for the solution on the defect ticket;
receiving a review result that indicates applying the solution failed to fix the defect;
creating new training data that includes the defect ticket, the solution, and the review result; and
retraining the correlation model using the new training data.

### Claim 12 (depends on 10)

The method of claim 10, further comprising:
generating a task to request review for the solution on the defect ticket;
receiving confirmation that the solution is a resolution for the defect;
creating new training data that includes the defect ticket, the solution, and the confirmation; and
retraining the correlation model using the new training data.

### Claim 13 (depends on 12)

The method of claim 12, further comprising:
indicating a resolve status on the defect ticket;
storing the defect ticket in the defects database; and
generating a user interface including a query tool for the defects database.

### Claim 14 (depends on 9)

The method of claim 9, the defect information indicating a high severity level and further comprising:
generating a high alert notification for the defect ticket based at least in part on the high severity level; and
pushing the high alert notification to at least one user account having a lead team role associated with the application.

### Claim 15 (depends on 9)

The method of claim 9, the event log associated with the error event being received in real-time or in near real-time, and further comprising:
determining, based at least in part on the corresponding defect information, a developer identifier associated with the error event and a stage of development cycle is associated with a development stage; and
pushing a high alert notification to at least a user account associated with the developer identifier.

### Claim 16

A method, comprising:
creating training data by identifying sample data from a defects database;
training a machine learning (ML) model with the training data to correlate input to identified defects;
receiving an event log;
determining, using the ML model, a defect associated with the event log correlates to an identified defect from the defects database;
generating a defect ticket for the defect with information including a solution of the identified defect indicated in the defects database;
receiving review results for applying the solution as a fix for the defect;
creating new training data including the defect ticket labeled with the review results; and
training a second ML model with the new training data.

### Claim 17 (depends on 16)

The method of claim 16, further comprising:
receiving an additional event log;
determining, using the second ML model, an additional defect associated with the additional event log fails to correlate to a second identified defect from the defects database; and
generating an additional defect ticket for the additional defect.

### Claim 18 (depends on 16)

The method of claim 16, further comprising:
receiving an additional event log;
determining, using the second ML model, an additional defect associated with the additional event log is a match for a second identified defect from the defects database;
retrieving a second defect ticket for the second identified defect from the defects database;
generating a log identifier for the additional event log; and
indicating the log identifier on the additional defect ticket.

### Claim 19 (depends on 16)

The method of claim 16, further comprising:
receiving an additional event log;
determining, using the second ML model, an additional defect associated with the additional event log correlates to a second identified defect from the defects database;
retrieving a second defect ticket for the second identified defect from the defects database;
determining a severity level of the additional defect is high based at least in part on the second defect ticket indicating a high severity level; and
generating an additional defect ticket for the additional defect indicating the high severity level.

### Claim 20 (depends on 19)

The method of claim 19, further comprising:
generating a high alert notification for first subscribers of events for a first application associated with the second identified defect and second subscribers of events for a second application associated with the additional defect.
