Web Resource Ranking via Quality Score Offset Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
New web resources often struggle to be properly ranked by search engines due to a lack of historical data, such as click-through-rate, even if they are highly relevant to search queries, leading to lower visibility in search engine result pages.
Innovation Solution
A computer-implemented method and system that generates a raw quality score for a web resource based on statistical features, compares it to a reference score associated with similar user traffic levels, and applies an offset adjustment formula to determine an adjusted quality score, which influences the resource's ranking in search engine result pages, ensuring it is positioned closer to its future traffic level.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional ranking algorithms are used that rely on historical data such as click-through-rate, then established web resources can be accurately ranked, but new web resources cannot be properly ranked even if they are highly relevant to search queries
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing quality scores for web resources based on their content and structural features before they receive user traffic. This allows new web resources to have initial ranking scores available immediately upon creation, without needing to wait for historical interaction data to accumulate. The quality score computation happens in advance during indexing, enabling new resources to be ranked from the first query they appear in.
Solution Approach 2:
The patent introduces an intermediary quality score metric that bridges the gap between content quality assessment and user interaction-based ranking. This quality score acts as a mediator that combines content analysis with predicted user engagement metrics, allowing the system to rank new resources based on their inherent quality and potential value rather than requiring actual historical performance data. The intermediary score translates content attributes into ranking-relevant metrics.
2Measurement precision
If search engines wait for sufficient historical data before ranking new web resources, then ranking accuracy improves, but the visibility and potential traffic of new resources decreases
Solution Approach 1:
The system performs preliminary quality assessment and score computation during the indexing phase, before any user interactions occur. This preliminary action ensures that when new web resources are first queried, they already have calculated quality scores that reflect their content value, allowing them to gain visibility and attract traffic immediately rather than waiting for historical data to accumulate.
Solution Approach 2:
The patent implements feedback mechanisms where actual user interactions (clicks, dwell time, returns) continuously update and refine the quality scores of web resources. This feedback loop allows the system to start with preliminary scores for new resources, then progressively improve accuracy as real user behavior data becomes available, balancing immediate visibility with ongoing ranking optimization.
3Reliability
If search engines apply strict quality thresholds based on historical performance, then high-quality established resources are well-ranked, but new resources with potential but no history are excluded
Solution Approach 1:
The patent applies parameter changes by transforming the ranking criteria from purely historical performance metrics to a hybrid model that includes content-based quality features, structural attributes, and predicted engagement metrics. This parameter transformation allows the system to assess new web resources using their inherent characteristics rather than requiring historical performance parameters, enabling potential resources to be identified and ranked based on their content quality and expected value.
Data Source
AI summary
There is disclosed a method and system for ranking a web resource. The method comprises: receiving a set of features associated with the web resource; generating a raw quality score based on the set of features, comparing the raw quality score to a reference score, the reference score being associated with a similar level of user traffic as the web resource; applying an offset adjustment formula to the raw quality score in response to an offset between the raw quality score and the reference score to obtain an adjusted quality score; in response to receiving a query, generating a search engine result page (SERP) including an indication of the web resource at a ranked position based on the adjusted quality score, the adjusted quality score causing a future raw quality score to be closer to a future reference score associated with a future level of user traffic.