Knowledge Graph Inference Path Confidence Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large-scale knowledge graphs are densely connected, leading to many irrelevant paths when querying, making current solutions costly and difficult to scale, and resulting in incorrect answers due to irrelevant paths.
Innovation Solution
A knowledge-scaling system that automatically learns relevant inference steps and confidence scores for inference paths within a knowledge graph structure, allowing for robust inferences and natural language explanations, and improves database lookup and keyword searches by refining query constraints and expanding search terms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all paths in a densely connected knowledge graph are followed for a given query, then comprehensive coverage is achieved, but computational cost increases and incorrect answers are returned due to irrelevant paths
Solution Approach 1:
The patent extracts and removes irrelevant paths from the knowledge graph traversal process. By identifying and eliminating paths that do not contribute to valid answers, the system maintains comprehensive coverage of relevant information while reducing computational waste and incorrect results from irrelevant path exploration
Solution Approach 2:
The patent changes parameters of path evaluation by introducing confidence scores and relevance metrics. Instead of treating all paths equally, the system dynamically adjusts the weight and consideration of different paths based on their confidence scores, allowing high-reliability paths to be prioritized while low-reliability paths are deprioritized or excluded
2Measurement precision
If manual construction of inference paths is performed, then precision of inference is improved, but time and effort requirements increase significantly
Solution Approach 1:
The patent performs preliminary action by pre-computing and storing confidence scores for inference paths in advance. This preliminary computation allows the system to have high-precision inference paths ready when needed, eliminating the need for time-consuming manual construction while maintaining precision through pre-analyzed confidence metrics
Solution Approach 2:
The system performs self-service by automatically learning and constructing inference paths with confidence scores without requiring manual intervention. The knowledge graph system itself generates and refines the inference paths through automated processes, reducing both manual effort and time while maintaining precision through iterative learning
3Quantity of substance
If knowledge graphs increase in size and complexity, then more comprehensive data is available, but scalability becomes difficult and computational cost increases
Solution Approach 1:
The patent applies parameter changes by introducing confidence scores as a filtering mechanism that scales with knowledge graph size. As the knowledge graph grows, the confidence score parameter allows the system to automatically prioritize high-quality paths and exclude low-quality ones, maintaining scalability despite increasing data volume and complexity
Data Source
AI summary
Aspects discussed herein present a solution for utilizing large-scale knowledge graphs for inference at scale and generating explanations for the conclusions. In some embodiments, aspects discussed herein learn inference paths from a knowledge graph and determine a confidence score for each inference path. Aspects discussed herein may apply the inference paths to the knowledge graph to improve database lookup, keyword searches, inferences, etc. Aspects discussed herein may generate a natural language explanation for each conclusion or result from one or more inference paths that led to that conclusion or result. Aspects discussed herein may present the best conclusions or results to the user based on selection strategies. The presented results or conclusions may include generated natural language explanations rather than links to documents with word occurrences highlighted.