BotDGT: Dynamicity-aware Social Bot Detection with Dynamic Graph Transformers

TL;DR

BotDGT tackles social bot detection by modeling social networks as dynamic graphs, combining a structural module for per-snapshot topology with a temporal module that aggregates historical context via temporal attention and position embeddings. The approach explicitly captures evolving behavior patterns and historical interaction context, addressing evasion strategies that static graph methods miss. Experiments on TwiBot-20 and TwiBot-22 show BotDGT outperforming static graph baselines in accuracy, recall, and F1-score, with ablations confirming the value of each component. The work demonstrates that incorporating dynamicity in graph representations can significantly improve detection of disguised bots and informs future developments in dynamic network anomaly detection.

Abstract

Detecting social bots has evolved into a pivotal yet intricate task, aimed at combating the dissemination of misinformation and preserving the authenticity of online interactions. While earlier graph-based approaches, which leverage topological structure of social networks, yielded notable outcomes, they overlooked the inherent dynamicity of social networks -- In reality, they largely depicted the social network as a static graph and solely relied on its most recent state. Due to the absence of dynamicity modeling, such approaches are vulnerable to evasion, particularly when advanced social bots interact with other users to camouflage identities and escape detection. To tackle these challenges, we propose BotDGT, a novel framework that not only considers the topological structure, but also effectively incorporates dynamic nature of social network. Specifically, we characterize a social network as a dynamic graph. A structural module is employed to acquire topological information from each historical snapshot. Additionally, a temporal module is proposed to integrate historical context and model the evolving behavior patterns exhibited by social bots and legitimate users. Experimental results demonstrate the superiority of BotDGT against the leading methods that neglected the dynamic nature of social networks in terms of accuracy, recall, and F1-score.

Figures (4)

• Figure 1: Dynamic nature of social network
• Figure 2: Overview of our proposed BotDGT framework.
• Figure 3: Framework performance at various time intervals
• Figure 4: Distribution of temporal attention weights

Theorems & Definitions (1)

• Definition : Dynamic Social Network