Transformed Social Network Graphs for Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social network graphs are affected by noise due to dissimilar users, leading to poor performance in search engines and advertising delivery systems, as weak connections between users with little in common can overstate similarity and introduce errors.
Innovation Solution
A transformed graph is generated by removing edges between nodes with fewer common neighbors and adding edges between nodes with more common neighbors, using a threshold-based approach to categorize and grow subgraphs, which helps in identifying more accurate user similarities and relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the social network graph includes all user relationships (edges), then the graph comprehensively represents the social network structure, but noise is introduced from weak connections between dissimilar users, reducing measurement precision
Solution Approach 1:
The patent extracts and removes weak edges from the social network graph based on a threshold criterion. Specifically, edges connecting users with fewer common neighbors (below a threshold value) are removed from the graph. This extraction process eliminates noisy weak connections while preserving strong connections, thereby improving user similarity measurement accuracy without losing comprehensive social network structure representation.
2Measurement precision
If edges are removed from the graph to reduce noise, then measurement precision improves, but the graph becomes less complete and may lose useful relationship information
Solution Approach 1:
The patent applies local quality by differentiating between strong and weak edges based on the number of common neighbors. Instead of uniformly treating all edges, the graph assigns different qualities to edges: strong edges (with sufficient common neighbors) are retained, while weak edges (with insufficient common neighbors) are removed. This localized differentiation improves measurement precision for user similarity while preserving the most meaningful relationship information.
3Adaptability or versatility
If the graph includes users with multiple independent relationships (multiplex connections), then the graph comprehensively represents diverse social connections, but dissimilar users connected through different relationship types introduce noise that overstates similarity
Solution Approach 1:
The patent uses common neighbors as an intermediary criterion to evaluate the strength of connections between users. Rather than directly measuring similarity based on multiple relationship types, the method introduces the intermediate metric of common neighbors. Users connected through multiple independent relationships will have more common neighbors, which serves as a mediator to distinguish strong meaningful connections from weak spurious connections, thereby improving similarity measurement accuracy while preserving adaptability to represent diverse relationship types.
Data Source
AI summary
A graph of a social network is received. The graph may include a node for each user account and an edge between nodes that represent social networking relationships such as messages between the user accounts or a friend relationship. The graph is transformed into a transformed graph where nodes have direct edges depending on a local test among its neighbors in the original graph. Small subsets of the transformed graph are categorized. The categories are used to identify subgraphs in the transformed graph. Each subgraph is grown by adding an edge from the transformed graph to the subgraph depending on local tests among nodes associated with the edge that have at least one edge that is already in the subgraph. The categorized subgraphs are used to provide targeted advertising, suggest new connections, identify different personalities and interests of users, or to provide other services.