Aggregated Social Feed Relevance Scoring and Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users accessing multiple social networking systems on mobile devices face impracticality due to limited display area and time constraints, as they need to sift through a large volume of irrelevant items from various feeds to find relevant content.
Innovation Solution
An aggregated social feed is created by selecting and scoring relevant items from multiple social networking system feeds based on relevance factors such as origin, interest level, popularity, and quality, and grouping criteria like subject, author, or geo-location, to reduce the velocity of items displayed, ensuring only the most relevant content is shown.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple social networking feeds are displayed on a mobile device, then the user receives comprehensive social updates, but the limited display area and user time constrain the ability to find relevant items
Solution Approach 1:
The patent extracts and removes duplicate or near-duplicate items from the aggregated feed before display. By identifying and eliminating redundant content across multiple social networking feeds, the system reduces the total volume of items users must review while preserving unique relevant content, thereby decreasing the time users spend searching for valuable information
Solution Approach 2:
The patent applies scoring mechanisms that dynamically adjust the visibility and prioritization of items based on multiple parameters including user preferences, item relevance, source credibility, and engagement metrics. This parameter-based filtering transforms the raw aggregated feed into a prioritized display that surfaces the most relevant items first, reducing the effective search time for users
2Quantity of substance
If duplicate or near duplicate items are removed from the feed, then the volume of items is reduced, but the complexity of processing and comparing items increases
Solution Approach 1:
The patent performs preliminary duplicate detection and removal operations on the aggregated feed data before it is prepared for display. By pre-processing the feed to identify and eliminate duplicates using content hashing and similarity algorithms, the system reduces the item volume early in the processing pipeline, preventing the burden of duplicate processing from compounding through subsequent display and interaction layers
3Measurement precision
If items are scored based on multiple relevance factors, then the accuracy of relevant item selection is improved, but the computational resources required increase
Solution Approach 1:
The patent implements a tiered scoring approach where items are evaluated using a subset of relevance factors based on their initial filtering stage and user profile characteristics. Not all items undergo complete multi-factor scoring - instead, the system applies computationally intensive relevance analysis selectively to items that pass preliminary filters, achieving high accuracy for critical items while conserving computational resources on less promising content
Data Source
AI summary
Relevant items are selected from personalized items included in a variety of social networking system feeds based on a relevance threshold. Content included in the received items is observed to facilitate grouping the items. Items satisfying a grouping criteria are grouped based on the content of the items. Items are then scored based on relevance factors, such as whether an item is included in the group or an indication of interest level associated with items in a group. Scored items meeting a relevance threshold are selected for display in an aggregated social feed in a content region of a page.