Knowledge Graph Triplet Expansion via Relationship Path Mining
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing knowledge graphs have limitations in expanding data and strengthening associations between entities, as they primarily focus on partial relationships within a local context, lacking the ability to provide a more global understanding.
Innovation Solution
A method and apparatus for expanding data in knowledge graphs by acquiring a triplet, mining a relationship path equivalent to the relationship in the triplet, and generating an expanded triplet, which strengthens the association between the subject and object in a larger context by replacing or traversing the relationship path.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a direct edge connects two entities in a knowledge graph, then the relationship is simple and easy to represent, but the association is partial and lacks global context
Solution Approach 1:
The patent segments a direct relationship edge into multiple intermediate relationship steps by introducing relationship paths. Instead of representing a relationship as a single direct edge, the system breaks it down into a sequence of smaller relationship hops through intermediate entities, thereby preserving global context information while maintaining manageable complexity at each step.
Solution Approach 2:
The patent introduces intermediate entities and relationship paths as mediators between subject and object entities. These intermediaries provide additional contextual information and global understanding of the relationship, transforming a simple direct connection into a enriched multi-hop path that captures more comprehensive semantic meaning.
2Loss of information
If relationship paths are mined and used to expand triplets, then the association between entities becomes more global and interpretable, but the data processing complexity increases
Solution Approach 1:
The patent performs preliminary action by pre-mining relationship paths from the knowledge graph before using them for triplet expansion. The system identifies and stores equivalent relationship paths in advance, creating a reusable resource that can be applied during the expansion process without performing complex path mining during inference, thereby reducing real-time processing complexity.
Solution Approach 2:
The patent uses copying by creating expanded triplets that replicate the structure of original triplets but with enriched relationship paths. Instead of modifying the original knowledge graph structure, the system generates copies of triplets with additional path information, preserving the original simple structure while providing enriched representations for analysis and training purposes.
3Productivity
If triplets are expanded using mined relationship paths, then the training sample set is enriched and model learning effectiveness is improved, but the computational resources and time required increase
Solution Approach 1:
The patent applies partial action by selectively expanding only certain triplets that benefit most from relationship path enrichment, rather than expanding all triplets uniformly. The system identifies high-value candidates for expansion based on criteria such as relationship complexity or importance, applying expansion efforts partially to achieve significant learning effectiveness improvement without the full computational cost of universal expansion.
Data Source
AI summary
A method, apparatus, device, and storage medium for expanding data are disclosed. The method includes: acquiring a triplet from a knowledge graph; mining a relationship path equivalent to a relationship in the triplet from the knowledge graph, a subject in the triplet being used as a start point of the relationship path, and an object in the triplet being used as an end point of the relationship path; and expanding the triplet based on the relationship path to generate an expanded triplet. This implementation expands the triplet in the knowledge graph, and strengthens the association between the subject and the object in the triplet in a larger context, such that the association between the subject and the object in the triplet is more global.