Knowledge Derivation System for Heterogeneous Data Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional knowledge derivation systems struggle with integrating information from multiple heterogeneous data sources, leading to segregated results that require users to manually sift through multiple sources, which is time-consuming and subjective, often missing relationships between data elements across sources.
Innovation Solution
A knowledge derivation system that extracts key entities from disparate data silos, normalizes and maps them using a dynamic domain ontology and semantic embeddings, automatically adjusts policies and relationships, and ranks results to provide integrated and relevant problem resolutions, independent of user skill level.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional knowledge derivation systems crawl multiple knowledge sources to retrieve resolutions, then comprehensive problem diagnosis information can be gathered, but the results become segregated and require manual sifting through multiple sources, increasing time consumption and subjectivity
Solution Approach 1:
The patent merges multiple segregated knowledge sources (user guides, technotes, forums, tickets) into a single unified knowledge base. The system extracts entities from each source, augments them with metadata and relationships, and integrates them together, allowing users to retrieve comprehensive problem diagnosis information from one consolidated source rather than manually searching across multiple segregated sources.
2Measurement precision
If users manually review multiple segregated results to retrieve relevant information, then comprehensive understanding can be achieved, but the process becomes highly time-consuming and subjective depending on user skill level
Solution Approach 1:
The system performs self-service by automatically extracting entities from multiple knowledge sources, augmenting them with metadata and relationships, and integrating them into a unified knowledge base. This automated process eliminates the need for users to manually review and synthesize information from multiple sources, while still providing comprehensive problem understanding through the enriched, interconnected data structure.
3Quantity of substance
If conventional systems return segregated results from multiple data sources, then all available information is provided, but relationships between data elements across sources are often missed
Solution Approach 1:
The patent introduces an intermediary entity augmentation process that acts as a mediator between segregated data sources. The system extracts entities from each source, then augments them with metadata, relationships, and contextual information from other sources. This intermediary augmentation layer connects previously isolated data elements, preserving relationships between data elements while maintaining the quantity of information from all sources.
4Adaptability or versatility
If multiple heterogeneous data sources are integrated, then a comprehensive knowledge base can be built, but the complexity of extracting and normalizing entities from disparate sources increases
Solution Approach 1:
The patent applies segmentation by breaking down the complex task of integrating heterogeneous data sources into distinct modular steps: entity extraction from each source, entity augmentation with metadata and relationships, and final integration into the unified knowledge base. This segmented approach manages the complexity of processing multiple heterogeneous sources by handling each source and transformation step separately and systematically.
Data Source
AI summary
Deriving augmented knowledge defining a knowledge base by extracting entities from a plurality of heterogeneous data sources; and augmenting the extracted entities; and utilizing an augmented entity to enhance a user activity.