Scalable and Generalizable Social Bot Detection through Data Selection

By Kai-Cheng Yang, Onur Varol, Pik-Mai Hui, Filippo Menczer. AAAI Conference on Artificial Intelligence, 2019

411 citations recorded by the source corpus.

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.

Subfields: Information Systems, Sociology and Political Science, Signal Processing

View Paper: https://www.semanticscholar.org/paper/7b6d4a3b0f713a3ddf8034b69f56bf9da678cc1b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Systems

In Sociology and Political Science

In Signal Processing

Citations in the corpus

A decade of social bot detection (2020, background): “A desirable scenario for the near future would involve the possibility to evaluate any new bot detector against many different types of social bots, thus moving along the y axis of Figure 7, following the promising approaches recently developed in [11, 36].”
Finding and Characterizing Information Warfare Campaigns (2020, background): “This is caused by the fact that supervised models can be brittle and are biased by their training data toward specific bot types and genres [37, 261].”