Social Network Connection Recommendations via Semantic and Relational Distance

Solution Overview

Problem

Existing social network systems primarily recommend connections based on relational closeness, limiting the introduction of new connections to users who already share a close bond, rather than suggesting individuals with shared interests or skills who may be relationally distant but meaningful.

Innovation Solution

A connection recommendation system that utilizes semantic analysis of user-generated content and measures of relational distance to suggest connections between users who share common interests or skills, even if they are not relationally close, by creating a lexical space that focuses on low polysemy lexical items and performing content clustering to identify potential connection candidates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If connection recommendations are based on relational closeness, then connection reliability is improved, but connection diversity deteriorates

Engineering Contradiction:

Improveconnection reliabilityVSAvoidconnection diversity

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system segments the connection recommendation process into two independent evaluation dimensions: relational distance (social graph proximity) and content similarity (lexical space clustering). By separating these criteria, the system can independently optimize for both reliability (through relational filtering) and diversity (through content-based candidate generation), then combine them to achieve balanced recommendations that neither prior art approach could accomplish alone.

Inventive Principle:

Principle #1Segmentation

2Adaptability or versatility

If connection recommendations are based on content similarity, then connection versatility is improved, but connection reliability deteriorates

Engineering Contradiction:

Improveconnection versatilityVSAvoidconnection reliability

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system segments the connection recommendation process into two independent evaluation dimensions: relational distance (social graph proximity) and content similarity (lexical space clustering). By separating these criteria, the system can independently optimize for both reliability (through relational filtering) and diversity (through content-based candidate generation), then combine them to achieve balanced recommendations that neither prior art approach could accomplish alone.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary filtering mechanism that uses relational distance as a gatekeeper for content-based recommendations. Content similarity identifies diverse candidates, but relational distance acts as an intermediary filter to ensure reliability, allowing only candidates within acceptable relational thresholds to proceed to final recommendation. This intermediary layer resolves the reliability-versus-versatility contradiction by mediating between the two competing criteria.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS10002127B2Connecting people based on content and relational distance

Publication Date: 2018.06.19 INTEL CORP

AI summary

Technologies for establishing connections between or among people based at least in part on semantic distance and relational distance include one or more computing devices that analyze content created by computing device users, perform content clustering on the content, determine relational distances between connection candidates, and generate connection recommendations based at least in part on the content clustering and the relational distances.