Unlearnable Graph: Protecting Graphs from Unauthorized Exploitation
This addresses privacy concerns for individuals and organizations using graph data by preventing unauthorized model training, representing a novel application in graph security.
The paper tackles the problem of unauthorized exploitation of graph data for training GNN models by proposing a method to generate unlearnable graphs, which reduces accuracy on the COLLAB dataset from 77.33% to 42.47% with at most 5% edge modifications.
While the use of graph-structured data in various fields is becoming increasingly popular, it also raises concerns about the potential unauthorized exploitation of personal data for training commercial graph neural network (GNN) models, which can compromise privacy. To address this issue, we propose a novel method for generating unlearnable graph examples. By injecting delusive but imperceptible noise into graphs using our Error-Minimizing Structural Poisoning (EMinS) module, we are able to make the graphs unexploitable. Notably, by modifying only $5\%$ at most of the potential edges in the graph data, our method successfully decreases the accuracy from ${77.33\%}$ to ${42.47\%}$ on the COLLAB dataset.