Tensor Model Decomposition for Social Network Latent Interaction Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current social network analysis tools are unable to simultaneously represent multiple types of relationships, topics of discussion, roles, properties of entities, and temporal states in a single representation, and lack the capability to visualize time, topics, and ranked importance of entities in social networks.
Innovation Solution
A method and system that process social network data using tensor models of at least order four, decompose them into principal factors, and synthesize a summary tensor to represent relationships among entities, identifying correlations, similarities, and time-based trends in relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If current social network analysis tools are used to represent multiple types of relationships, topics, roles, properties, and temporal states, then the representation becomes comprehensive and unified, but the device complexity and computational requirements increase significantly
Solution Approach 1:
The patent applies tensor decomposition to transform complex multi-dimensional social network data into a series of simpler two-dimensional matrices. By decomposing the tensor into factors and reconstructing them as matrices, the system reduces computational complexity while preserving the ability to represent multiple relationship types, topics, and temporal states simultaneously.
Solution Approach 2:
The patent segments the complex tensor data into multiple principal factors through decomposition. Each factor can be independently processed and interpreted, allowing the system to handle complex social network relationships by breaking them down into manageable components that can be analyzed separately before being synthesized back into a comprehensive representation.
2Loss of information
If social network analysis tools incorporate time, topics, and ranked importance of entities, then the analysis becomes more insightful and comprehensive, but the ease of operation and visualization decreases
Solution Approach 1:
The patent transforms multi-dimensional tensor data containing time, topic, and importance information into two-dimensional matrix representations. This dimensional reduction simplifies the data structure while preserving all critical information, making it easier to visualize and operate with while maintaining comprehensive analytical capabilities.
Solution Approach 2:
The patent creates simplified matrix copies from the original complex tensor data through decomposition. These matrix representations serve as manageable copies that retain the essential information about time patterns, topics, and entity importance, allowing analysts to work with simpler structures that are easier to interpret and visualize.
3Measurement precision
If the tensor model is decomposed into multiple principal factors, then the measurement precision of latent interactions improves, but the computational time and processing resources increase
Solution Approach 1:
The patent applies partial decomposition by identifying and extracting only the most significant principal factors from the tensor. Rather than decomposing into all possible factors, the system determines an optimal number of factors to retain that provides sufficient precision for detecting latent interactions, thereby reducing unnecessary computational overhead while maintaining measurement accuracy.
Data Source
AI summary
A method including processing social network data to establish a tensor model of the social network data, the tensor model having at least an order of four. The tensor model is decomposed into a plurality of principal factors. A summary tensor is synthesized from a subset of the plurality of principal factors. The summary tensor represents a plurality of relationships among a plurality of entities in the tensor model. A synthesis of relationships is formed and stored. At least one parameter is identified using one of the summary tensor and a single principal factor in the subset. The at least one parameter is selected from the group consisting of: a correlation among the plurality of entities, a similarity between two of the plurality of entities, and a time-based trend of changes in the synthesis of relationships. The at least one parameter is communicated.