Native Application Score Normalization for Search Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search engines struggle to effectively rank native applications and web page resources relative to each other, leading to potential underserving or penalization of high-quality resources due to disparate scoring systems and lack of cross-attribution of relevance and quality signals.
Innovation Solution
A system that receives and normalizes web resource scores and native application scores to a common scale, using global and user device-specific signals, and applies a normalization factor to adjust native application scores, allowing for cross-attribution of relevance and quality signals between web page resources and native applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If native application scores and web resource scores are used separately without normalization, then each score type can be calculated independently using its own scoring system, but the scores cannot be fairly compared or ranked together due to disparate scoring scales
Solution Approach 1:
The patent transforms native application scores and web resource scores from disparate scoring scales to a common normalized scale through mathematical transformation. A normalization factor is calculated based on the relationship between native application scores and corresponding web resource scores, then applied to convert all scores to a comparable range, enabling fair ranking while maintaining the independence of original scoring systems.
Solution Approach 2:
The patent introduces a normalization factor as an intermediary element that mediates between native application scores and web resource scores. This normalization factor acts as a bridge, translating scores from different scoring systems into a common language that allows for meaningful comparison and unified ranking without requiring complete integration of the underlying scoring mechanisms.
2Measurement precision
If cross-attribution of relevance and quality signals is implemented between native applications and web resources, then the accuracy of relevance assessment is improved, but the computational complexity and data processing requirements increase
Solution Approach 1:
The patent merges relevance signals and quality signals from both native applications and corresponding web resources into a unified assessment framework. By combining these signals through cross-attribution, the system leverages the strengths of both score types to produce a more accurate overall relevance assessment, where the normalization process integrates the combined signals into a coherent ranking metric.
3Productivity
If normalization is applied to all native application scores, then comparability between native applications and web resources is achieved, but high-quality web resources with low-quality corresponding native applications may be unfairly penalized
Solution Approach 1:
The patent applies local quality adjustment by recognizing that not all native application-web resource pairs have equal quality relationships. The normalization process selectively adjusts scores based on the specific relationship between each native application and its corresponding web resource, allowing high-quality web resources to maintain their integrity even when their corresponding native applications are of lower quality, rather than applying a blanket normalization that would unfairly penalize them.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium for normalizing native application scores. In an aspect, a system ranks web resources and native applications based on web resource scores and normalized native application scores that are normalized to the web resource scores. The ranking is indicative of the relevance of each web resource and native application for a search operation relative to each other web resource and native application.