Summary of A General Black-box Adversarial Attack on Graph-based Fake News Detectors, by Peican Zhu et al.
A General Black-box Adversarial Attack on Graph-based Fake News Detectors
by Peican Zhu, Zechen Pan, Yang Liu, Jiwei Tian, Keke Tang, Zhen Wang
First submitted to arxiv on: 24 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes the General Attack via Fake Social Interaction (GAFSI) framework, a first-of-its-kind approach that targets Graph Neural Network (GNN)-based fake news detectors in black-box scenarios. The proposed method simulates sharing behaviors to fool these detectors, which rely on graph structures to learn distinctive news embeddings for classification. Specifically, GAFSI involves a fraudster selection module and a post injection module to select engaged users and create shared relations, respectively. Experimental results on empirical datasets demonstrate the effectiveness of GAFSI. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Fake news detection is an important issue in today’s digital world. Researchers have been using Graph Neural Networks (GNNs) to detect fake news by analyzing graph structures. However, these detectors can be fooled if attackers know how they are constructed. This paper proposes a new way for attackers to trick GNN-based fake news detectors without knowing the specific details of their construction. The method is called General Attack via Fake Social Interaction (GAFSI). It simulates sharing behaviors on social media to create fake relationships that can fool these detectors. The authors tested this method on some datasets and it worked. |
Keywords
» Artificial intelligence » Classification » Gnn » Graph neural network
