Dynamic Threshold Data Acquisition for Search Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data acquisition methods for promotional platforms are static and inflexible, leading to poor-quality search results and a worse correlation between displayed products and search terms, impacting user experience and click-through rates due to the use of uniform threshold scores that fail to adapt to dynamic changes in correlation threshold scores.
Innovation Solution
A dynamic data acquisition system that uses a threshold value dictionary to assign dynamic threshold scores based on real-time textual and data analysis characteristic factors, allowing for flexible and accurate filtering of product data information, which adapts to changes in click-through rates and user behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If fixed threshold values are used for filtering product data information, then the filtering process is simple and fast, but the search result quality and correlation with search terms deteriorate
Solution Approach 1:
The patent implements dynamic threshold values that automatically adjust based on real-time correlation scores between search terms and product data information. Instead of using fixed thresholds, the system calculates correlation scores dynamically and sets thresholds adaptively, allowing the filtering mechanism to respond to changing data relationships while maintaining operational efficiency
Solution Approach 2:
The system changes the threshold parameter from a static fixed value to a dynamic value that varies based on correlation analysis. The threshold is no longer a constant but is derived from real-time calculations of the relationship between search terms and product information, enabling the filtering criteria to adapt to different search contexts and data patterns
2Ease of manufacture
If uniform threshold scores are applied to all search terms, then the filtering method is simple to implement, but the adaptability to dynamic changes in correlation scores deteriorates
Solution Approach 1:
The patent transforms the static uniform threshold score into a dynamic threshold that automatically adapts to changing correlation relationships. The system continuously calculates correlation scores between search terms and product data, then adjusts thresholds accordingly, enabling the filtering mechanism to respond to dynamic changes without requiring complex manual reconfiguration
Solution Approach 2:
The system implements self-adjusting thresholds that automatically respond to changes in data correlations without external intervention. The threshold values are derived from the system's own correlation analysis, allowing the filtering mechanism to self-optimize based on the actual relationships between search terms and product information
3Productivity
If ad search results are displayed before natural search results, then promotional products are highlighted, but user experience and click-through rates deteriorate
Solution Approach 1:
The patent applies different display strategies to different types of search results based on their quality and relevance. High-quality, highly correlated product information is prioritized in prominent positions, while less relevant promotional content is displayed in secondary positions. This local differentiation of display quality ensures that the most relevant information receives the most attention
4Measurement precision
If correlation threshold scores are calculated and stored in a database, then the correlation between keywords and product information is captured, but the system complexity and data processing overhead increase
Solution Approach 1:
The patent pre-calculates and stores correlation threshold scores in a database for efficient retrieval during search operations. By performing the correlation analysis in advance and caching the results, the system avoids repeating complex calculations for every search query, thereby reducing real-time processing overhead while maintaining accurate correlation measurements
Data Source
AI summary
Acquiring dynamic data is disclosed including extracting a search term from a search request string that is received, looking up the search term in a threshold value dictionary to acquire a dynamic threshold score corresponding to the search term, using the search term as a query condition and the dynamic threshold score corresponding to the search term as a filter condition to acquire, in an index data table, one or more corresponding pieces of index information, acquiring data information corresponding to the search term based on the index information in the index data table, and sending the data information to be displayed in a page of a website. The dynamic threshold score varies based on a characteristic factor.