Social Media Stance Detection and Clustering for Narrative Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional processes for analyzing social media data are inefficient and lack sufficient context, often requiring manual analysis of large datasets and producing results that do not accurately distinguish between supportive and critical narratives, while automated methods fail to provide comprehensive insights into the spread of information and misinformation across networks.

Innovation Solution

A computerized method involving attentive clustering and knowledge graph embeddings is employed to analyze social media data, generating clusters and metrics such as contagion scores, homophily, and heterophily, and using stance detection to understand the spread and influence of content across platforms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual analysis of social media data is performed, then contextual understanding and accuracy in distinguishing supportive and critical narratives is improved, but time consumption and processing efficiency deteriorate

Engineering Contradiction:

Improveaccuracy in distinguishing supportive and critical narrativesVSAvoidtime consumption for analyzing large datasets

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent introduces stance detection models as intermediary components that automatically analyze social media content to determine supportive, critical, or neutral positions. These models serve as mediators between raw data and human analysts, providing pre-processed insights that maintain accuracy while reducing time consumption. The system uses multiple layers of automated analysis including entity recognition, sentiment analysis, and relationship extraction to preserve contextual understanding without requiring manual review of every data point.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces manual mechanical analysis with automated computational systems. Machine learning models and natural language processing algorithms substitute human analysts for routine classification tasks, automatically identifying narrative positions and relationships in social media data. This substitution maintains measurement precision through sophisticated algorithms while dramatically reducing the time required to process large datasets.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If automated methods are used to analyze social media data, then processing efficiency and speed are improved, but contextual understanding and ability to distinguish between supportive and critical narratives deteriorate

Engineering Contradiction:

Improveprocessing efficiency of social media dataVSAvoidcontextual understanding of narratives

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The patent implements nested analytical layers where multiple analysis functions are embedded within each other. The system nests entity recognition within sentiment analysis, which is in turn nested within stance detection, which is nested within broader narrative analysis. This nested structure allows automated processing at each layer while preserving contextual information through the hierarchy, enabling high productivity without losing nuanced understanding of supportive and critical narratives.

Inventive Principle:

Principle #7Nested doll (Nesting)

Solution Approach 2:

The patent dynamically adjusts analysis parameters based on data characteristics and analysis depth requirements. The system can modify parameters such as analysis granularity, confidence thresholds, and contextual window sizes to optimize both processing efficiency and contextual understanding. By changing parameters adaptively, the system maintains high productivity while preserving necessary contextual information for accurate narrative distinction.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If comprehensive automated analysis is implemented, then productivity and speed of analysis are improved, but system complexity and computational resources required deteriorate

Engineering Contradiction:

Improvespeed of social media data analysisVSAvoidsystem complexity for automated analysis

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the comprehensive analysis system into modular functional components including data collection modules, preprocessing modules, entity recognition modules, sentiment analysis modules, stance detection modules, and visualization modules. Each module performs a specific function independently, allowing the system to achieve high productivity through specialized processing while managing complexity through modular design. This segmentation enables independent optimization of each component and simplifies system maintenance and deployment.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS12499160B2Analyzing social media data to identify markers of coordinated movements, using stance detection, and using clustering techniques

Publication Date: 2025.12.16 GRAPHIKA TECH INC

AI summary

The present disclosure relates to methods for analyzing social media data using various techniques, including classifying at least one contagious phenomenon propagating on a network, providing stance detection based on the social media data, clustering the social media data, and/or generating and analyzing knowledge graph embeddings using the social media data.