Intent-Driven Knowledge Graph Notifications for Relevant Updates

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing alert systems for relevant information suffer from 'goldilocks syndrome', where users receive either too much or too little information, making it difficult to identify important updates, leading to missed information or time inefficiencies.

Innovation Solution

A user-defined intent is registered to monitor a knowledge graph for relevant changes, with a listener triggering alerts when updates match the intent, and a natural language generator processes change information into easily understandable messages.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If a user subscribes to all relevant newsletters, then the user receives complete information coverage, but the user becomes overwhelmed and cannot read all newsletters

Engineering Contradiction:
Improveinformation coverageVSAvoiduser workload
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The system segments the broad category of 'all relevant newsletters' into specific knowledge graph elements (nodes and edges) that match the user's defined intent. Instead of presenting all newsletters uniformly, the system divides information into targeted updates based on the user's specific interests in the knowledge graph, allowing complete coverage without overwhelming the user.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The knowledge graph serves as an intermediary between the user and the information sources. The user defines intent at the knowledge graph level, and the system uses the knowledge graph structure to automatically filter and translate this into specific newsletter subscriptions and change notifications, eliminating the need for the user to manually manage multiple newsletter subscriptions.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If a user subscribes to a subset of newsletters, then the user reduces information overload, but the user may miss important information

Engineering Contradiction:
Improveuser workloadVSAvoidinformation coverage
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The system implements feedback through the knowledge graph structure, where changes to specific nodes or edges automatically trigger notifications. The system continuously monitors the knowledge graph for changes that match the user's intent and provides feedback in the form of targeted alerts, ensuring important information is delivered without requiring the user to subscribe to multiple newsletters.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The user performs preliminary action by defining their intent at the knowledge graph level before information changes occur. This intent definition pre-configures the system to automatically identify and notify the user of relevant changes, eliminating the need for the user to manually select individual newsletters while ensuring complete information coverage.

Inventive Principle:
Principle #10Preliminary action

3Loss of information

If a user scans all newsletters manually, then the user ensures complete information review, but the user spends excessive time on scanning

Engineering Contradiction:
Improveinformation review completenessVSAvoidtime spent scanning
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The system replaces the manual mechanical process of scanning newsletters with an automated computational system. The knowledge graph monitoring and change detection mechanisms automatically identify and notify users of relevant changes, substituting human time and effort with automated processes that provide complete information review without time expenditure by the user.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs self-service by automatically monitoring the knowledge graph for changes and generating notifications without user intervention. The system independently identifies relevant changes based on the user's defined intent and delivers information updates autonomously, eliminating the need for manual scanning while ensuring complete information review.

Inventive Principle:
Principle #25Self-service

4Loss of information

If a user receives detailed change information, then the user gets comprehensive update details, but the user has difficulty identifying the most relevant information

Engineering Contradiction:
Improveinformation detail completenessVSAvoidrelevance identification
Core Design Contradiction:
Loss of informationVSDifficulty of detecting and measuring

Solution Approach 1:

The system applies local quality by tailoring the information delivery to the specific user intent defined at the knowledge graph level. Instead of providing uniform detailed information for all changes, the system selectively provides detailed notifications only for changes that locally match the user's specific interests, making relevant information easily identifiable while maintaining completeness.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP4617900A1Customized information change notification using knowledge graphs
Publication Date: 2025.09.17 SAP SE
  • EP4617900A1 patent drawingFigure 1~2
  • EP4617900A1 patent drawingFigure 3
  • EP4617900A1 patent drawingFigure 4

AI summary

Techniques and solutions are provided for providing alerts to users when information changes, particularly information associated with a knowledge graph. A user can define an intent, where the intent describes the type of information for which a user desires to receive alerts. The intent can be specified directly with respect to knowledge graph elements, or the intent can be specified in another manner and mapped to such elements. A listener is implemented for the intent. A knowledge graph is periodically reviewed for updates. Updates that are relevant to a particular user intent cause the associated listener to be triggered, and information regarding the update is then provided to the user.