Search Result Segmentation Using User Relationship Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search engines face challenges in providing sufficient and relevant personalized search results due to limited user information, leading to inadequate user experience from excessive personalization or insufficiently personalized results.
Innovation Solution
A method and apparatus that combine user relationship data with search queries to enhance search results, involving the acquisition of user relationship data and integrating it with search results, using a recommendation model to rank and annotate results, thereby providing more comprehensive and personalized search outcomes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If completely personalized search services are provided using user information, then search results are personalized according to user preferences, but users may be unable to acquire information really needed due to excessive personalization
Solution Approach 1:
The patent segments search results into two distinct parts: universal search results and personalized search results. Universal search results ensure comprehensive information coverage, while personalized search results provide tailored recommendations. This segmentation allows the system to maintain both broad information availability and personalized relevance without one compromising the other.
Solution Approach 2:
The patent applies local quality by providing different types of search results for different user needs. Rather than uniformly personalizing all results, the system selectively personalizes certain results while maintaining universal results for general information needs. This allows different parts of the search result set to serve different functions - some for comprehensive coverage and others for personalized relevance.
2Ease of operation
If only simple user information such as gender, age, and personal preference is used for personalization, then the search system is easy to operate, but the acquired personalized search results may be insufficient and cannot satisfy the users' need for personalized search
Solution Approach 1:
The patent introduces user relationship data as an intermediary that bridges the gap between simple user information and high-quality personalization. Instead of directly using complex user behavior data or requiring extensive user input, the system leverages relationship data (such as social connections, interaction patterns) that naturally reflects user preferences and needs. This intermediary data source enables sophisticated personalization while maintaining ease of operation, as the relationship data is automatically collected rather than requiring direct user input.
3Measurement precision
If user relationship data is integrated with search queries, then the relevance and abundance of personalized search results are enhanced, but the system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-processing and organizing user relationship data before it is needed for search operations. The system establishes relationship data structures, defines relationship types, and prepares the data in advance, so that during actual search operations, the integration with queries becomes a straightforward process rather than requiring complex real-time computations. This preliminary preparation reduces the complexity burden during runtime search operations.
Data Source
AI summary
A search request including user information and query information sent by a user is received. A search is conducted according to the query information to acquire a first search result associated with the query information. User relationship data associated with the user information is acquired. A second search result associated with both the query information and the user relationship data is acquired. A display order for various search result items in the first search result and the second search result is set. The first search result and the second search result are sent to the user after the display order is set. The techniques of the present disclosure provide abundant personalized search data for users, and meanwhile avoid poor user experience caused by excessive personalization.