Application Object Label Management for Real-Time ML Data Tagging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing software applications face challenges in integrating machine learning (ML) and artificial intelligence (AI) due to the time-consuming and resource-intensive process of adding labels during data preparation, leading to delays between data updates and model usage, especially in SaaS environments where upgrades require manual adjustments.

Innovation Solution

A Label Management System (LMS) that enables in-application, real-time labeling of data objects and related objects, allowing users to extend and enrich data with custom labels, and automates label computation using rules and large language models, ensuring minimal re-labeling and lifecycle stability across software updates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If data labeling is performed during data preparation pipeline, then ML model training can be performed, but time delay occurs between data updates and model usage

Engineering Contradiction:
Improvetime delay between data updates and model usageVSAvoidspeed of ML model training
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The system performs preliminary labeling actions by adding label extension fields to application objects and pre-configuring labeling rules within the software application itself. This allows label data to be prepared in advance as part of the application object structure, eliminating the need for separate post-export labeling processes and reducing the time delay between data updates and model training.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent merges the data labeling functionality directly into the software application by integrating label extension fields and labeling rules within the application's data structure. This combines previously separate processes (data extraction, labeling, and model training) into a unified system where labeling occurs inline with application object management, improving productivity by eliminating intermediate steps.

Inventive Principle:
Principle #5Merging (Combining)

2Reliability

If manual labeling is performed during data engineering, then complete label data can be obtained, but resource consumption increases

Engineering Contradiction:
Improvecompleteness of label dataVSAvoidresource consumption for data preparation
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system enables self-service labeling by allowing the software application to automatically generate and maintain label data through integrated labeling rules and extension fields. The application objects themselves carry their label information, eliminating the need for external manual labeling processes and reducing resource consumption while maintaining data completeness.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent introduces label extension fields as an intermediary structure that bridges application objects and ML label requirements. These extension fields serve as a mediator that stores label data directly within the application object structure, allowing automatic population of label information without requiring separate manual labeling processes, thus reducing resource consumption while ensuring complete label data.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Ease of manufacture

If data is exported for external labeling, then labeling can be performed, but integration complexity increases

Engineering Contradiction:
Improveease of data labelingVSAvoidintegration complexity of labeling system
Core Design Contradiction:
Ease of manufactureVSDevice complexity

Solution Approach 1:

The system creates a universal labeling framework where label extension fields and labeling rules are integrated directly into the software application's data structure. This multi-functional approach allows the same application object structure to serve both application logic and ML labeling requirements, eliminating the need for separate export/import processes and reducing integration complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4700659A1Software application label management for machine learning
Publication Date: 2026.02.25 SAP SE
  • EP4700659A1 patent drawingFigure 1
  • EP4700659A1 patent drawingFigure 2
  • EP4700659A1 patent drawingFigure 3

AI summary

A computer-implemented method includes adding, using a label management system (LMS), an object-specific multi-value label extension field to an application object (AO) associated with a software application. Using the LMS and to an AO user interface (UI) associated with the AO, a multi-value label UI extension field is added. Using the LMS, a label value table and associated label value maintenance UI is created to maintain label values. Using the LMS, a mapping table is created between the object-specific multi-value label extension field of the AO and the label value table. Using the LMS and the AO UI, label values assigned to the AO are read using foreign key retrieval from the mapping table and the label value table.