Knowledge Graph Augmentation via Attribute Statistical Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional knowledge graph augmentation methods rely on additional sources or perform basic label-based analysis, failing to efficiently augment existing knowledge graphs by not considering the content or values of attributes and focusing primarily on tables rather than attributes.
Innovation Solution
A system and method for knowledge graph augmentation using statistical analysis of attributes, involving mapping classes, attributes, and instances, indexing semantically similar data elements through label-based, content-based, and attribute-based clustering, and ranking them to create a ranked list.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional techniques use label-based analysis only, then the system is simple to implement, but the augmentation efficiency is insufficient
Solution Approach 1:
The patent combines label-based analysis with value-based analysis into a unified framework. The system simultaneously processes both the structural labels (column headers) and the actual data values to identify relevant attributes, thereby improving augmentation efficiency while maintaining implementation feasibility through a integrated approach.
Solution Approach 2:
The patent creates a multi-functional analysis system that performs both label-based matching and value-based statistical analysis. This universal approach allows the system to handle diverse data sources and attribute types effectively, improving overall augmentation efficiency without requiring separate specialized systems.
2Device complexity
If conventional techniques focus on tables rather than attributes, then the system structure is straightforward, but the attribute content and values are not adequately considered
Solution Approach 1:
The patent transitions from a table-centric single-dimension approach to an attribute-centric multi-dimensional approach. By analyzing attributes across multiple dimensions (labels, values, statistical properties), the system recovers information about attribute content and values that would be lost in traditional table-focused methods.
Solution Approach 2:
The patent changes the focus parameter from table structure to attribute characteristics. By computing statistical measures (frequency, entropy, correlation) of attribute values and analyzing value distributions, the system captures detailed attribute content information that was previously overlooked in table-structured approaches.
3Adaptability or versatility
If conventional techniques rely on additional external sources, then the knowledge graph can be augmented with external information, but the system complexity and dependency on external sources increases
Solution Approach 1:
The patent enables the knowledge graph augmentation system to serve itself by extracting and analyzing attribute information directly from the input structured data. The system computes all necessary statistical measures and identifies relevant attributes using only the provided data, eliminating dependencies on external sources while maintaining effective augmentation capability.
Data Source
AI summary
A method, system, and recording medium for knowledge graph augmentation using data based on a statistical analysis of attributes in the data, including mapping classes, attributes, and instances of the classes of the data, indexing semantically similar input data elements based on the mapped data using at least one of a label-based analysis, a content-based analysis, and an attribute-based clustering, and ranking the semantically similar input data elements to create a ranked list.