Cognitive Platform for Knowledge Extraction from Heterogeneous Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in generating effective knowledge graphs from heterogeneous data sources, as they require holistic knowledge models that capture unique enterprise information and relations, but struggle to extract and infer relevant data from structured and unstructured sources efficiently.
Innovation Solution
A method and cognitive platform for knowledge extraction that uses pre-existing knowledge structures to identify concepts and relations from heterogeneous data sources, mapping entities to concepts and relations, and converting these into data structures to create a second knowledge structure, enabling the association of data sources and enhancing knowledge harvesting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a holistic knowledge model is created to capture unique enterprise information, then the quality and comprehensiveness of knowledge extraction is improved, but the complexity of the knowledge model and data processing increases
Solution Approach 1:
The knowledge extraction process is divided into multiple stages: concept extraction from unstructured data, relation identification, entity extraction from structured data, and mapping to knowledge graph. This segmentation allows the complex task to be handled through manageable steps, each with specific algorithms and data types, reducing the overall complexity while maintaining comprehensive knowledge capture
Solution Approach 2:
The system performs preliminary concept extraction and relation identification from unstructured data sources before processing structured data. This preliminary action establishes a knowledge framework that guides subsequent entity extraction and mapping operations, improving extraction quality while managing complexity through preparatory organization
2Quantity of substance
If multiple heterogeneous data sources are processed to harvest comprehensive knowledge, then the completeness of knowledge graph is improved, but the time and resources required for data processing increases
Solution Approach 1:
The system extracts concepts and identifies relations from unstructured data sources as a preliminary step, creating a structured knowledge framework before processing structured data sources. This preliminary processing reduces the complexity of subsequent mapping operations and enables more efficient integration of multiple data sources
Solution Approach 2:
The patent introduces intermediate data structures and mapping layers that serve as mediators between different heterogeneous data sources and the final knowledge graph. These intermediaries standardize the representation of entities and relations, enabling efficient processing and integration across diverse data formats without requiring direct complex interactions between all data sources
Data Source
AI summary
Provided is a method and platform for knowledge extraction from multiple data sources. The disclosure includes extracting entities from the data sources and synthesizing the same. They are then classified into concepts or connectors. Based on the extracted data and identified entities, knowledge model is created. The knowledge model can show the relation between the concept and the connectors. Once the knowledge model is created, a second data source is used. Using the knowledge model and second data source, data records are created which can be used to prepare the knowledge graph.