Association User List Updates for Real-Time Recommendations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing user recommendation systems face high calculation costs and poor real-time performance due to the need to determine user associations from large datasets, making real-time user push impossible.
Innovation Solution
An information processing method that dynamically updates an association user list in response to a trigger instruction, determining and displaying second users associated with a first user based on a server recommendation model and real-time adjustments, reducing calculation load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If user association is determined from large quantities of users using existing pushing manner, then comprehensive user recommendation can be achieved, but calculation amount becomes large and real-time performance deteriorates
Solution Approach 1:
The patent segments the user recommendation process into two distinct phases: offline pre-computation phase where association relationships are pre-calculated and stored in an association relationship graph, and online real-time phase where only graph traversal is performed. This segmentation allows comprehensive user recommendation to be achieved through pre-computation while enabling real-time performance during actual pushing operations.
Solution Approach 2:
The patent performs preliminary computation by pre-calculating association relationships between users offline and storing them in an association relationship graph structure. This preliminary action ensures that when real-time user pushing is needed, the system only needs to traverse the pre-built graph rather than computing associations from scratch, thus achieving both comprehensive recommendations and real-time performance.
2Reliability
If user association is determined from large quantities of users using existing pushing manner, then comprehensive user recommendation can be achieved, but calculation cost becomes high
Solution Approach 1:
The patent segments the computation into offline pre-computation (performed once or periodically) and online real-time querying (performed frequently). By separating these phases, the high calculation cost is amortized over time, while maintaining comprehensive and accurate user recommendations through the pre-built association relationship graph.
Solution Approach 2:
The system performs preliminary computation of user associations offline and stores results in an optimized graph structure. This preliminary action eliminates the need for repeated expensive calculations during real-time operations, significantly reducing calculation cost while preserving recommendation accuracy.
3Ease of operation
If association user list is updated in real-time based on user actions, then user experience is improved, but calculation amount increases
Solution Approach 1:
The patent implements a dynamic association user list that automatically updates based on user actions (such as following or unfollowing). When dynamic events occur, the system efficiently updates the association relationship graph and triggers real-time pushes to affected users, improving user experience while managing calculation through event-driven updates rather than continuous recomputation.
Solution Approach 2:
The system establishes a feedback mechanism where user actions (following/unfollowing) trigger automatic updates to the association user list and real-time pushes. This feedback loop ensures the recommendation system adapts to user behavior in real-time, improving ease of operation while controlling calculation through targeted updates based on specific user events.
Data Source
AI summary
Provided are an information processing method, an electronic device, and a storage medium. The method includes the following: in response to a trigger instruction from a current user for a first user in an association user list, the association user list is dynamically updated and displayed; where the association user list corresponds to the current user, and the step in which the association user list is dynamically updated includes that at least one second user is determined according to the first user, and the association user list is updated based on the at least one second user; where the at least one second user is associated with the first user.