Search Site Quality Classification via Link Score Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Search engines face challenges in distinguishing and demoting low-quality sites in search results, leading to users being presented with irrelevant information.
Innovation Solution
A system that calculates a link quality score for sites by analyzing the distribution of resource quality scores from resources linking to the site, classifies sites as low quality if the score is below a threshold, and adjusts the ranking scores of search results accordingly to prioritize higher quality sites.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If search engines return all candidate search results, then the quantity of search results is maximized, but the quality and relevance of results deteriorates due to inclusion of low-quality sites
Solution Approach 1:
The patent applies parameter changes by introducing a link quality score threshold parameter. Resources are filtered based on whether their link quality score meets or exceeds this threshold, transforming the search result selection from a quantity-maximizing approach to a quality-filtered approach. This parameter-based filtering resolves the contradiction by allowing the system to maintain high result quantities while ensuring quality through the threshold mechanism.
Solution Approach 2:
The patent implements local quality by evaluating and scoring individual resources based on their specific link quality metrics rather than treating all search results uniformly. Each resource receives a link quality score based on analysis of linking resources, and this localized quality assessment enables selective inclusion in search results, thereby maintaining both quantity and quality.
2Reliability
If search engines decrease ranking scores of low-quality sites, then the quality of search results is improved, but the complexity of the ranking system increases
Solution Approach 1:
The patent applies preliminary action by calculating link quality scores for resources before they are included in search results. The system pre-assesses each resource's link quality based on analyzing linking resources and assigns a score in advance. This preliminary quality assessment integrates seamlessly into the existing ranking system without requiring complex real-time modifications, thus improving result quality while minimizing additional system complexity.
3Measurement precision
If search engines analyze link quality scores of resources, then the precision of quality assessment is improved, but the time and computational resources required increase
Solution Approach 1:
The patent extracts the quality assessment function into a separate, dedicated process that calculates link quality scores independently of the main search query processing. By separating the quality assessment (analyzing linking resources and calculating scores) from the search result generation, the system achieves precise quality measurement without significantly increasing the time required for actual search operations. The extracted assessment can be performed in advance or in parallel.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for enhancing search results. In one aspect, a method includes receiving a resource quality score for each of a plurality of resources linking to a site/ Each of the resources is assigned to one of a plurality of resource quality groups, each resource quality group being associated with a range of resource quality scores, each resource being assigned to the resource quality group associated with the range encompassing the resource quality score for the resource. The number of resources in each resources quality group is counted. A link quality score is determined for the site using the number of resources in each resource quality group. If the link quality score is below a threshold link quality score, the site is classified as a low quality site.