ML Knowledge Graph for Supply Chain Relationship Discovery

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current systems face challenges in efficiently identifying and quantifying supply chain relationships between companies, particularly from unstructured data sources, which are vast and complex, making it difficult for non-technical users to generate analytics and for financial institutions to make informed investment decisions.

Innovation Solution

A system utilizing natural language processing and machine learning to analyze unstructured text data, building knowledge graphs that integrate with existing structured data, and providing a natural language interface for users to query supply chain relationships, enabling the computation of confidence scores for supplier-customer relations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If machine learning and natural language processing are used to analyze unstructured text data, then the ability to automatically identify and quantify supply chain relationships is improved, but the system complexity and computational resources required increase

Engineering Contradiction:

Improveautomatic identification of supply chain relationshipsVSAvoidsystem complexity

Core Design Contradiction:

Extent of automationVSDevice complexity

Solution Approach 1:

The patent employs natural language processing and machine learning algorithms as intermediary components that automatically process unstructured text data to extract supply chain relationships. These intermediaries bridge the gap between raw unstructured data and structured relationship information, enabling automatic identification without manual intervention while managing complexity through specialized processing layers.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system replaces manual mechanical analysis of unstructured text with automated computational processes using machine learning and natural language processing. This substitution transforms the complex task of manually extracting supply chain relationships from unstructured data into an automated computational pipeline, significantly improving automation extent while managing complexity through algorithmic approaches.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If knowledge graphs are built to integrate structured and unstructured data, then the quality of analytics and risk assessment is improved, but the time and computational resources required for data processing increase

Engineering Contradiction:

Improvequality of analyticsVSAvoiddata processing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary processing of unstructured text data by pre-processing documents to extract entities and relationships before integrating them into the knowledge graph. This preliminary action prepares data in advance for more efficient integration and querying, improving analytics quality while reducing the time required during actual analysis operations by having data ready in structured formats.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The knowledge graph construction process is segmented into distinct stages: extracting entities from unstructured text, identifying relationships between entities, and integrating this information into the structured knowledge graph. This segmentation allows each processing stage to be optimized independently, improving overall analytics quality while managing time consumption through efficient modular processing.

Inventive Principle:

Principle #1Segmentation

3Ease of operation

If natural language interfaces are provided for querying supply chain relationships, then the ease of use for non-technical users is improved, but the complexity of the query processing system increases

Engineering Contradiction:

Improveease of use for queryingVSAvoidquery processing complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The natural language interface acts as an intermediary layer that translates user-friendly queries into complex structured queries against the knowledge graph. This intermediary handles the complexity of query processing internally while presenting a simple interface to users, improving ease of operation without exposing the underlying system complexity to end users.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11222052B2Machine learning-based relationship association and related discovery and

Publication Date: 2022.01.11 REFINITIV US ORGANIZATION LLC

AI summary

Systems and techniques for determining relationships and association significance between entities are disclosed. The systems and techniques automatically identify supply chain relationships between companies based on unstructured text corpora. The system combines Machine Learning models to identify sentences mentioning supply chain between two companies (evidence), and an aggregation layer to take into account the evidence found and assign a confidence score to the relationship between companies.