Attribute-Tagged Search Recommendations for More Relevant Query Suggestions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of search recommendation words presented to users on search intermediate pages is low, failing to meet user search needs.
Innovation Solution
A recommendation word determination method that presents search results under different attribute tags and generates a recommendation word set based on the search word and attribute tag, improving relevance to user search needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If search recommendation words are presented on search intermediate pages, then users can search for relevant information, but the accuracy of search recommendation words is low
Solution Approach 1:
The patent applies dynamics by making the search recommendation words adaptive and changeable based on user interactions. The system dynamically adjusts recommendation words according to user search history, clicked categories, and real-time search behavior, transforming static recommendation words into dynamic ones that evolve with user needs, thereby improving accuracy without sacrificing search efficiency
Solution Approach 2:
The patent implements feedback mechanisms by monitoring user interactions with search results and recommendation words. User feedback including search history, category selections, and search result views is collected and fed back into the recommendation system to continuously optimize accuracy. This closed-loop feedback ensures recommendation words become more accurate over time while maintaining high search productivity
2Measurement precision
If search recommendation words are generated based on search word and attribute tag, then relevance to user search needs is improved, but system complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the recommendation word generation process into distinct modules: search word analysis module, attribute tag identification module, and recommendation word generation module. Each module handles specific tasks independently, making the complex system more manageable. The segmentation allows complex relevance optimization through modular components that can be developed, tested, and maintained separately
Solution Approach 2:
The patent utilizes parameter changes by adjusting multiple parameters including search word frequency, attribute tag weights, user behavior patterns, and time-based factors to optimize recommendation relevance. By changing and combining these parameters in different ways, the system achieves high relevance without requiring a single complex algorithm, instead using coordinated parameter adjustments across multiple processing stages
Data Source
AI summary
The present disclosure relates to a recommendation word determination method and apparatus, and an electronic device and a storage medium. The method comprises: receiving a search word, and presenting, on a search result page, a first search result corresponding to the search word under a first attribute tag, and an attribute of the first search result is matched with the first attribute tag; in response to a switching operation for the attribute tags, presenting, a second search result corresponding to the search word under a second attribute tag, wherein an attribute of the second search result is matched with the second attribute tag; and in response to a triggering operation for a search box, presenting a recommendation word set on a search intermediate page, the recommendation word set determined based on the search word and the second attribute tag. The search experience of a user is improved.