Automatic Text Mining Service Mapping via Taxonomy Instance Generation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current text mining services face challenges in accuracy and completeness due to differences in taxonomies used for annotating entities, relations, and classifying text documents, leading to difficulties in effectively combining and integrating results from multiple services.

Innovation Solution

The system automates the process of mapping different annotation taxonomies using an instance generation component and an auto-mapping component, which generates instances for taxonomy elements and computes mappings between taxonomies, allowing for the integration of results from multiple text mining services through a result combiner.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If multiple text mining services are used to extract information, then the quantity and coverage of extracted information is improved, but the accuracy and reliability of results deteriorate due to taxonomy differences and inconsistencies

Engineering Contradiction:
Improvequantity of extracted informationVSAvoidaccuracy of extracted information
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent introduces an intermediary component (taxonomy mapping service) that mediates between multiple text mining services with different taxonomies and the final result aggregation. This intermediary automatically maps entity types from different taxonomies to a unified taxonomy, enabling consistent integration of results while maintaining the benefits of multiple services. The mapping service acts as a translator that resolves taxonomy conflicts without requiring manual intervention.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system dynamically changes the taxonomy parameter by automatically selecting and applying appropriate mappings based on the specific text mining services being combined. The taxonomy mapping is not fixed but adapts to different combinations of services, allowing the system to optimize the unified taxonomy structure for each specific integration scenario, thereby maintaining accuracy while combining multiple services.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If manual mapping of taxonomies is performed to integrate text mining services, then the accuracy of result integration is improved, but the time and effort required deteriorates

Engineering Contradiction:
Improveaccuracy of result integrationVSAvoidtime for mapping process
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system implements self-service by enabling automatic taxonomy mapping without requiring manual intervention. The taxonomy mapping service automatically analyzes entity types from different text mining services, identifies correspondences, and creates mappings autonomously. This self-organizing capability eliminates the need for manual taxonomy alignment while maintaining high integration accuracy, thereby resolving the contradiction between accuracy and time consumption.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent performs preliminary taxonomy mapping analysis by examining service metadata and entity type definitions before actual result integration. This preliminary action prepares the mapping framework in advance, so that when results need to be integrated, the mapping is already established. This proactive approach significantly reduces the time required for integration while ensuring accurate mapping relationships are in place.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If text mining services are combined without mapping, then the productivity and speed of information extraction is improved, but the completeness and consistency of results deteriorate

Engineering Contradiction:
Improvespeed of information extractionVSAvoidcompleteness of extracted information
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The taxonomy mapping service serves as an intermediary that operates in parallel with the text mining services, maintaining the high productivity of direct extraction while ensuring completeness through automatic mapping. The mediator component processes entity types asynchronously and creates unified taxonomy mappings without blocking the primary information extraction workflow, thus preserving speed while improving completeness.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP2645309B1Automatic combination and mapping of text-mining services
Publication Date: 2016.11.30 SAP SE
  • EP2645309B1 patent drawingFigure 1
  • EP2645309B1 patent drawingFigure 1A
  • EP2645309B1 patent drawingFigure 1B

AI summary

Embodiments of systems and methods for automatic combination of text mining services, may comprise an instance generation component and an auto-mapping component. From common text sources, the instance generation component generates instances for taxonomy elements attached to a particular text mining service. These instances are then forwarded to the auto-mapping component, which computes a mapping between different taxonomies. This mapping may be saved to avoid repeated instance generation and mapping processes for similar taxonomies. The computed mapping may in turn be input to a result combiner element, together with extraction results from calling the different text mining services. The result combiner returns the merged result sets to the execution, and finally to the user or API.