Multiple Rumor Source Detection with Graph Convolutional Networks
Ming Dong, Bolong Zheng, Quoc Viet Hung Nguyen +2 more
Detecting rumor source in social networks is one of the key issues for defeating rumors automatically. Although many efforts have been devoted to defeating online rumors, most of them are proposed based an assumption that the underlying propagation model is known in advance. However, this assumption may lead to impracticability on real data, since it is usually difficult to acquire the actual underlying propagation model. Some attempts are developed by using label propagation to avoid the limitation caused by lack of prior knowledge on the underlying propagation model. Nonetheless, they still suffer from the shortcoming that the node label is simply an integer which may restrict the prediction precision. In this paper, we propose a deep learning based model, namely GCNSI (Graph Convolutional Networks based Source Identification), to locate multiple rumor sources without prior knowledge…