Scalable and Generalizable Social Bot Detection through Data Selection

TL;DR

This paper introduces a scalable, data-efficient social bot detection framework that uses minimal metadata and strategic data selection to improve accuracy, generalization, and interpretability in real-time Twitter analysis.

Contribution

It proposes a novel data selection approach for training social bot detectors, enhancing scalability, generalization, and interpretability over existing models.

Findings

Strategic data subset selection improves model accuracy.

Model generalizes well to unseen datasets.

Framework operates efficiently on real-time Twitter data.

Abstract

Efficient and reliable social bot classification is crucial for detecting information manipulation on social media. Despite rapid development, state-of-the-art bot detection models still face generalization and scalability challenges, which greatly limit their applications. In this paper we propose a framework that uses minimal account metadata, enabling efficient analysis that scales up to handle the full stream of public tweets of Twitter in real time. To ensure model accuracy, we build a rich collection of labeled datasets for training and validation. We deploy a strict validation system so that model performance on unseen datasets is also optimized, in addition to traditional cross-validation. We find that strategically selecting a subset of training data yields better model accuracy and generalization than exhaustively training on all available data. Thanks to the simplicity of the proposed model, its logic can be interpreted to provide insights into social bot characteristics.