Search Result Ranking via Entity Metrics and Knowledge Graph
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search result ranking techniques, such as alphabetical ordering and keyword matching, fail to effectively identify and prioritize the most relevant data, leading to suboptimal search outcomes.
Innovation Solution
A method and system that utilize a knowledge graph to determine a set of metrics, including relatedness, notable entity type, contribution, and prize metrics, which are weighted based on entity types to calculate a score for ranking search results, thereby enhancing relevance and presentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional techniques such as alphabetical ordering and keyword matching are used for ranking search results, then the implementation is simple, but the ability to effectively identify and prioritize relevant data deteriorates
Solution Approach 1:
The patent segments the search result ranking process into multiple independent metric components (e.g., entity type metric, contribution metric, prize metric, relatedness metric). Each metric is calculated separately based on specific criteria, and then combined through weighted summation. This segmentation allows the system to maintain high relevance assessment accuracy while managing complexity through modular design, where each metric can be independently optimized and adjusted.
2Measurement precision
If multiple weighted metrics are calculated from the knowledge graph to rank search results, then the relevance assessment improves, but the computational complexity increases
Solution Approach 1:
The patent applies local quality by determining entity types and assigning different weights to different metrics based on the specific entity type being evaluated. For example, certain metrics may be weighted more heavily for academic publications versus commercial products. This allows the system to optimize computational resources by focusing calculations on the most relevant metrics for each entity type, improving relevance scoring accuracy without uniformly increasing computational complexity across all search results.
Data Source
AI summary
Methods, systems, and computer-readable media are provided for ranking search results. A search system may determine several metrics based on search results. The search system may determine weights for the metrics, wherein the weights are based in part on the type of entity included in the search. The search system may determine a score by combining the metrics and the weights. The search system may rank search results based on the score.