Social-Media Activity Forecasting with Exogenous Information Signals

TL;DR

This paper introduces a forecasting model for social media activity that leverages external signals like news and conflicts, along with platform data, to predict daily topic-specific activity levels, aiding in strategic information analysis and intervention.

Contribution

The paper presents a novel modeling approach combining exogenous and endogenous data for accurate social media activity forecasting across multiple platforms and contexts.

Findings

Effective prediction of social media activity using combined signals

Demonstrated robustness across different platforms and topics

Improved understanding of information spread dynamics

Abstract

Due to their widespread adoption, social media platforms present an ideal environment for studying and understanding social behavior, especially on information spread. Modeling social media activity has numerous practical implications such as supporting efforts to analyze strategic information operations, designing intervention techniques to mitigate disinformation, or delivering critical information during disaster relief operations. In this paper we propose a modeling technique that forecasts topic-specific daily volume of social media activities by using both exogenous signals, such as news or armed conflicts records, and endogenous data from the social media platform we model. Empirical evaluations with real datasets from two different platforms and two different contexts each composed of multiple interrelated topics demonstrate the effectiveness of our solution.