A General Language for Modeling Social Media Account Behavior

TL;DR

This paper introduces BLOC, a flexible language for modeling social media account behaviors that improves detection of malicious activities like social bots and inauthentic campaigns without extensive fine-tuning.

Contribution

The paper presents BLOC, a novel general language for representing social media behaviors, enhancing detection accuracy across various malicious activities.

Findings

BLOC achieves comparable or better detection performance than state-of-the-art methods.

The language effectively models a broad spectrum of behaviors without extensive fine-tuning.

BLOC is adaptable to different types of social media account behaviors.

Abstract

Malicious actors exploit social media to inflate stock prices, sway elections, spread misinformation, and sow discord. To these ends, they employ tactics that include the use of inauthentic accounts and campaigns. Methods to detect these abuses currently rely on features specifically designed to target suspicious behaviors. However, the effectiveness of these methods decays as malicious behaviors evolve. To address this challenge, we propose a general language for modeling social media account behavior. Words in this language, called BLOC, consist of symbols drawn from distinct alphabets representing user actions and content. The language is highly flexible and can be applied to model a broad spectrum of legitimate and suspicious online behaviors without extensive fine-tuning. Using BLOC to represent the behaviors of Twitter accounts, we achieve performance comparable to or better than state-of-the-art methods in the detection of social bots and coordinated inauthentic behavior.