Personalized Location Recommendations via Social Influence Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing location-based discovery platforms fail to personalize recommendations to users' preferences, as they do not effectively utilize social network connections and generated content to tailor suggestions.
Innovation Solution
A system and method that utilize a server communicating with user devices to retrieve social network connection content, calculate influence scores for locations based on user preferences, and provide personalized location-based recommendations by synthesizing the influence of primary and secondary social network connections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If location-based discovery platforms provide general venue recommendations, then users can discover new locations, but the recommendations are not personalized to user preferences
Solution Approach 1:
The patent segments the social network into primary connections (direct friends/family) and secondary connections (friends of friends), allowing the system to process influence scores in hierarchical layers. This segmentation enables personalized recommendations by analyzing content from different relationship levels separately and combining their influences, achieving adaptability without overwhelming system complexity
Solution Approach 2:
The system pre-calculates influence scores for each social network connection based on their generated content, preferences, and behaviors before needing to make recommendations. These pre-computed influence scores are stored and reused when generating location recommendations, allowing the system to provide personalized results efficiently without recalculating everything from scratch each time
2Measurement precision
If the system analyzes social network connections and content to personalize recommendations, then recommendation accuracy improves, but data processing requirements increase
Solution Approach 1:
The patent extracts only the essential elements needed for personalization: influence scores from primary and secondary connections, and their associated preferences. Rather than processing all social network data, the system selectively extracts relevant influence metrics and preference information, achieving high recommendation accuracy while managing data processing volume efficiently
Solution Approach 2:
The system focuses on analyzing content and preferences from a targeted subset of social connections (those with highest influence scores) rather than uniformly processing all connections. By applying partial action to the most influential connections, the system achieves high recommendation accuracy without the computational burden of analyzing every single social network connection in equal detail
Data Source
AI summary
A method for providing a recommendation to a user, including retrieving, connection content associated with social network connections of the user; calculating, for each of the social network connections, an influence score for each of a plurality of locations; receiving a request location from a user device associated with the user; extracting a recommendation from relevant connection content, the relevant connection content being content associated with the request location and generated by at least one of the social network connections having the highest influence scores for the request location, the recommendation comprising a venue referenced within the relevant connection content; and sending the recommendation to the user device.