Emotional Sequential Influence Modeling on False Information

TL;DR

This paper introduces EUSIM, a model that captures how false information spreads emotionally among social media users, predicting future emotional responses based on social context and past interactions.

Contribution

It proposes the first emotionally infused sequential influence model, EUSIM, to analyze and predict emotional propagation patterns related to false information.

Findings

EUSIM effectively models emotional influence in social networks.

The model predicts users' future emotions with high accuracy.

Emotional influence depends on social context, content, and historical interactions.

Abstract

The extensive dissemination of false information in social networks affects netizens social lives, morals, and behaviours. When a neighbour expresses strong emotions (e.g., fear, anger, excitement) based on a false statement, these emotions can be transmitted to others, especially through interactions on social media. Therefore, exploring the mechanism that explains how an individuals emotions change under the influence of a neighbours false statement is a practically important task. In this work, we systematically examining the publics personal, interpersonal, and historical emotional influence based on social context, content, and emotional based features. The contribution of this paper is to build an emotionally infused model called the Emotional based User Sequential Influence Model(EUSIM) to understand users temporal emotional propagation patterns and predict future emotions against false information.