Network Prebunking Problem: Optimizing Prebunking Targets to Suppress the Spread of Misinformation in Social Networks

TL;DR

The paper tackles the spread of misinformation in social networks by formulating the network prebunking problem: selecting a budget-constrained set of prebunking targets to minimize misinformation diffusion before exposure. It proves NP-hardness and submodularity of the objective, enabling greedy guarantees, and introduces MIA-NPP, a scalable approximation based on Maximum Influence Arborescence that operates on local propagation trees. Through real-world datasets, it demonstrates that MIA-NPP consistently suppresses misinformation better than baselines and remains robust under parameter uncertainty. The work highlights prebunking as a feasible, ethically favorable strategy that complements existing blocking and clarification approaches, with practical implications for large-scale social platforms.

Abstract

As a countermeasure against misinformation that undermines the healthy use of social media, a preventive intervention known as \textit{prebunking} has recently attracted attention in the field of psychology. Prebunking aims to strengthen individuals' cognitive resistance to misinformation by presenting weakened doses of misinformation or by teaching common manipulation techniques before they encounter actual misinformation. Despite the growing body of evidence supporting its effectiveness in reducing susceptibility to misinformation at the individual level, an important open question remains: how best to identify the optimal targets for prebunking interventions to mitigate the spread of misinformation in a social network. To address this issue, we formulate a combinatorial optimization problem, called the \textit{network prebunking problem}, which aims to select optimal prebunking targets that minimizes the spread of misinformation in a social network under limited intervention budgets. We show that the problem is NP-hard and that its objective function is monotone and submodular, which provides a theoretical foundation for approximation guarantees of greedy algorithms. However, since the greedy algorithm is computationally expensive and does not scale to large networks, we propose an efficient approximation algorithm, MIA-NPP, based on the Maximum Influence Arborescence (MIA) approach, which restricts influence propagation around each node to a local directed tree rooted at that node. Through numerical experiments using real-world social network datasets, we demonstrate that MIA-NPP effectively suppresses the spread of misinformation under both fully observed and uncertain model parameter settings.

Figures (10)

• Figure 1: Schematic diagram of the network prebunking problem.
• Figure 2: Graph $H$ corresponding to a set cover instance with $U=\{u_1, \dots, u_5\}$ and $\mathcal{T}=\{\{u_1, u_2, u_3\}, \{u_2, u_4\}, \{u_3, u_4\}, \{u_4, u_5\}\}$.
• Figure 3: Misinformation suppression effects of each algorithm on the PolitiFact and GossipCop networks.
• Figure 4: Visualizations of the first 100 prebunking targets selected by each algorithm on the GossipCop network, plotted on a 2D plane of $pp(P_{rv}^\ast)$ and $q_v$. Nodes are shown as black dots, with selected targets marked as red Xs.
• Figure 5: Misinformation suppression effects of MIA-NPP under uncertain observations on the PolitiFact and GossipCop networks. Red and yellow dashed lines respectively indicate the results of MIA-NPP and AdvancedGreedy under perfect observations, for comparison.

Theorems & Definitions (3)

• theorem 1
• lemma 1
• theorem 2