Social Attribute Ascription for Content Feed Personalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social network services face challenges in providing personalized and relevant content to users, as existing content feeds often display a mix of content that may not align with individual user interests or attributes, leading to a lack of engagement and relevance.
Innovation Solution
A system that acquires and ascribes social attributes to content items based on user interactions, allowing for the selection and display of content that matches user profile attributes, such as job titles, education, skills, and location, to create highly customized and relevant content feeds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If content feeds display a mix of content in reverse chronological order, then the system maintains simplicity in content delivery, but the content relevance to individual user interests deteriorates
Solution Approach 1:
The patent applies local quality by customizing content feeds for different users based on their individual profile attributes. Instead of a uniform content delivery approach, the system analyzes user-specific characteristics (job title, education, skills, location) and selectively presents content that matches each user's profile, thereby maintaining simplicity in the overall system while improving content relevance for each individual user.
2Loss of information
If the system ascribes social attributes to content items based on user interactions, then content personalization improves, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing social attributes for content items based on user interactions before these items are displayed in feeds. The system analyzes user-content interactions, determines relevant social attributes (such as trending among specific user groups), and stores this information in advance. This allows the feed generation process to simply retrieve and display pre-analyzed content rather than performing complex analysis in real-time, thus improving personalization while managing system complexity.
3Reliability
If content feeds display trending content among similar users, then user engagement improves, but the complexity of content selection increases
Solution Approach 1:
The patent applies copying by creating and maintaining user profile attribute copies that can be efficiently compared and matched. Instead of performing complex real-time analysis of user similarities, the system stores simplified representations of user attributes and content characteristics, allowing for rapid matching and selection of trending content among similar users. This approach improves user engagement by displaying relevant trending content while reducing the computational complexity of the content selection process.
Data Source
AI summary
Techniques for ascribing social attributes to content items and for selecting content to display in a content feed are described. According to various embodiments, accessing one or more content items accessible via a network are accessed, each of the content items having received one or more social activity signals. Thereafter, members of an online social network service that submitted the social activity signals may be identified. Member profile data identifying member profile attributes of the members cemented the social activity signals may then be accessed. Thereafter, social attribute information may be generated and associated with each of the content items, the social attribute information identifying the member profile attributes of the members that submitted the social activity signals associated with each of the content items.