G-CutMix: A CutMix-based graph data augmentation method for bot detection in social networks
Authors:
Open Access
- 26 September 2025
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 20 (9) , e0331978
Abstract
The CutMix technique is a sophisticated approach for augmenting data in order to train neural network-based image classifiers. Essentially, it involves cutting out a portion of a random image and pasting it into the same location as another image. However, because of the irregularity of graph data, CutMix cannot be directly applied to graph learning. Our paper introduces G-CutMix, a CutMix-based data augmentation approach that we designed specifically for bot detection in social media networks. G-CutMix involves conducting CutMix operations between the original graph and a shuffled graph, which precedes the graph convolution process. The outputs of the graph convolution are then strategically merged with the user representations from both the original and shuffled graphs. Our proposed G-CutMix not only leverages the power of graph convolutions but also introduces a layer of complexity that mimics real-world scenarios where bot behavior can be subtle and varied, making G-CutMix a formidable tool in the arsenal against bot detection. Our experiments confirm that our approach can consistently enhance the performance of bot detection across various GNN architectures, including Graph Convolutional Networks, GraphSAGE, and Graph Attention Networks.Keywords
Funding Information
- Wuxi University Research Start-up Fund for High-level Talents
This publication has 25 references indexed in Scilit:
- Neighborhood Difference-Enhanced Graph Neural Network Based on Hypergraph for Social Bot Detection2024
- Bot2Vec: A general approach of intra-community oriented representation learning for bot detection in different types of social networksInformation Systems, 2022
- TrustGCN: Enabling Graph Convolutional Network for Robust Sybil Detection in OSNs2020
- CutMix: Regularization Strategy to Train Strong Classifiers With Localizable Features2019
- Modeling Relational Data with Graph Convolutional Networks2018
- SybilSCAR: Sybil detection in online social networks via local rule based propagation2017
- Fame for sale: Efficient detection of fake Twitter followersDecision Support Systems, 2015
- Twitter spammer detection using data stream clusteringInformation Sciences, 2014
- Empirical Evaluation and New Design for Fighting Evolving Twitter SpammersIEEE Transactions on Information Forensics and Security, 2013
- SybilLimit: A Near-Optimal Social Network Defense against Sybil Attacks2008 IEEE Symposium on Security and Privacy (SP 2008), 2008