Cross Session Diversity Scoring for Social Networking Feeds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social networking systems face challenges in prioritizing and diversifying activity feed content for users across multiple sessions, often resulting in repetitive content from the same sources, which can lead to user engagement decline and dissatisfaction.
Innovation Solution
The system generates profile value scores and discount factors for activity feed items based on user interactions and social graph data, ensuring that content from diverse sources is prioritized by calculating relevance scores and adjusting display frequencies of items from frequent contributors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system prioritizes content from frequent contributors based on traditional relevance algorithms, then the activity feed shows highly relevant content, but content diversity decreases and repetition increases
Solution Approach 1:
The patent applies parameter changes by introducing a discount factor that dynamically adjusts the relevance score based on the frequency of display. The relevance score is modified from a static calculation to a dynamic one that incorporates temporal frequency information, transforming the scoring parameter to balance relevance and diversity across multiple sessions
Solution Approach 2:
The system implements feedback by tracking past session display data and using it to adjust current session relevance scores. The discount factor is calculated based on feedback from previous sessions' display history, creating a closed-loop system that continuously optimizes content selection to prevent repetition while maintaining relevance
2Reliability
If the system displays the same content repeatedly to frequent contributors, then user engagement with familiar content increases, but user dissatisfaction grows due to repetition
Solution Approach 1:
The patent applies preliminary anti-action by pre-calculating discount factors based on projected repetition patterns before content is displayed. The system anticipates the harmful effect of repetition and counteracts it in advance by reducing the relevance score of content that has been displayed frequently in past sessions, preventing user dissatisfaction before it occurs
3Productivity
If the system uses simple relevance scoring without considering past session data, then computation time is reduced, but cross-session content diversity is lost
Solution Approach 1:
The system applies preliminary action by pre-computing and storing discount factors based on past session data before the current session begins. The relevance scoring algorithm uses these pre-calculated factors, which are derived from historical display patterns, to quickly adjust content ranking without requiring real-time analysis of entire session histories, thus maintaining speed while improving cross-session diversity
Data Source
AI summary
A system and method for personalizing cross session diversity is disclosed. The system receives a member opportunity request. In response, the system generates a list of members in response to the received member opportunity request, wherein the list of members is determined based on member profile data stored at a social networking system. For each member in the generated list of members, the system generates a profile value score based on the stored member profile data. The system ranks the members of the generated list at least in part based on the generated profile value scores. The system then selects one or more members in the list of members based on the ranking of members in the generated list.