Dynamic Search Result Re-ranking via Integrated Evaluation Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Push information in search results often appears in a fixed position and has lower relevance to search words compared to natural search results, leading to a lower click rate due to poor integration of evaluation scores.
Innovation Solution
A method that integrates evaluation scores from natural search and push information lists by re-ranking natural search materials based on combined scores, optimizing score weights according to user experience indices, and updating push evaluation parameters to improve relevance and positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If push information is placed in a fixed position in search results, then information pushing is achieved, but relevance between push information and search word deteriorates
Solution Approach 1:
The patent implements dynamic positioning of push information by calculating integrated evaluation scores that combine natural search relevance and push information scores. The position of push information in search results is no longer fixed but dynamically adjusted based on the integrated score, allowing the system to maintain both pushing functionality and relevance quality.
Solution Approach 2:
The patent changes the parameter of push information positioning from a fixed position to a position determined by integrated evaluation scores. By introducing the score integration mechanism with weighting parameters, the system can flexibly adjust the ranking position of push information based on its relevance to the search word, thus resolving the contradiction between pushing efficiency and relevance.
2Ease of operation
If push information is integrated with natural search results, then user experience is improved, but evaluation score integration complexity increases
Solution Approach 1:
The patent segments the evaluation score integration process into distinct components: natural search evaluation scores and push information evaluation scores. Each component is calculated separately using its own service (natural search service and push information search service), and then integrated through a weighted sum formula. This segmentation reduces integration complexity by maintaining independent evaluation mechanisms while achieving unified ranking.
3Measurement precision
If re-ranking is performed based on integrated evaluation scores, then search result accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary actions by pre-calculating and caching natural search evaluation scores and push information evaluation scores before the final ranking stage. The natural search service and push information search service independently compute their respective scores in advance, which are then integrated and used for re-ranking. This preliminary computation reduces the time required for final re-ranking while maintaining accuracy.
Data Source
AI summary
A method of pushing information is provided. According to an example, a search server obtains a natural search list including a first threshold number of natural search materials and a first evaluation score of each of the natural search materials by invoking a natural search service to perform a search operation on a search word, obtains a push information list including a second threshold number of push materials and a second evaluation score of each of the push materials by invoking a push information search service to perform a search operation on a search word, integrates the first evaluation score and the second evaluation score of a material existing in the natural search list and the push information list at the same time, re-ranks the natural search materials according to integrated evaluation scores, and pushes re-ranked natural search materials.