Machine-Learned Error Notification Filtering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for notifying users of errors in electronic apparatuses often result in low importance notifications being frequent, leading to users discounting important alerts and potentially missing critical information.

Innovation Solution

An information processing device that employs a machine-learned model to determine when to notify users of errors based on error information, operation information, and user actions, using a neural network to learn and optimize notification conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If error notifications are made frequently to ensure users are informed of all errors, then information completeness is improved, but user attention and notification effectiveness deteriorate due to information overload

Engineering Contradiction:

Improveerror information completenessVSAvoiduser attention and notification effectiveness

Core Design Contradiction:

Loss of informationVSEase of operation

Solution Approach 1:

The patent changes the parameter of notification selection from static user-defined rules to dynamic machine learning-based importance assessment. The system learns from historical user actions (whether users accessed notification details) to dynamically adjust which error notifications are sent, transforming the notification decision from a fixed parameter to an adaptive one that optimizes both information completeness and user attention.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent implements feedback loops where user responses to notifications (accessing details or ignoring) are collected and fed back into the machine learning model. This feedback mechanism allows the system to continuously improve its understanding of user priorities and adjust notification strategies accordingly, resolving the contradiction between providing complete information and maintaining user attention.

Inventive Principle:

Principle #23Feedback

2Ease of operation

If users manually set notification conditions to filter important errors, then notification relevance is improved, but system complexity and user burden increase

Engineering Contradiction:

Improvenotification relevanceVSAvoidsystem complexity and user burden

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent applies self-service by enabling the system to automatically learn and determine notification conditions without requiring manual user configuration. The machine learning model autonomously analyzes user behavior patterns and error characteristics to generate optimized notification strategies, eliminating the need for users to manually set complex filtering conditions while maintaining high notification relevance.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical system of manual rule-setting with an intelligent machine learning system. Instead of users manually configuring notification parameters based on their understanding of error importance, the system uses algorithms to automatically learn and apply optimal notification conditions, substituting human cognitive effort with automated intelligent processing.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Loss of information

If all error notifications are sent to users, then information completeness is improved, but user burden and information discounting worsen

Engineering Contradiction:

Improveerror information completenessVSAvoiduser burden

Core Design Contradiction:

Loss of informationVSEase of operation

Solution Approach 1:

The patent applies partial action by selectively sending only the most important error notifications to users based on machine learning assessments of error importance and user behavior patterns. Rather than sending all error notifications (excessive action), the system sends a curated subset that maximizes information value while minimizing user burden, achieving the optimal balance between information completeness and user experience.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS11310371B2Information processing device, learning device, and non-transitory recording medium storing machine-learned model

Publication Date: 2022.04.19 SEIKO EPSON CORP

AI summary

An information processing device includes a storage configured to store a machine-learned model, a reception section configured to receive error information and operation information transmitted from an electronic apparatus, and a processor configured to determine whether a notification indicating error information is to be made for a user. The machine-learned model mechanically learns a condition of the notification indicating error information to be made to the user based on a data set in which the error information, the operation information, and action information indicating user action performed in response to the notification of the error information are associated with one another.