[Objective] Improve the accuracy of real-time detection of rumors in social media and reduce the harm caused by rumors.

[Methods] A real-time rumor detection method based on graph attention neural network integrated with knowledge graph is proposed. First, obtaining the background knowledge of the text from the external knowledge graph through knowledge distillation. Second, transforming the text and background knowledge into a weighted graph structure representation through point mutual information, and a weighted graph attention neural network is used to learn the discontinuous semantic features of the text from the weighted graph. Then, learning the continuous semantic features of the text through the pre-trained language model BERT, and the statistical features of users and content are converted into continuous vector representations using the embedding method. Finally, all the features are fused and input into the fully connected neural network for rumor detection.

[Results] Experimental results on two public social media rumor datasets, PHEME and WEIBO, show that the method's accuracy reaches 92.1% and 84.0%, respectively, higher than the state-of-the-art baseline methods.

[Limitations] The method does not fuse the image or video information that may be attached to the post and cannot perform multi-modal fusion rumor detection.

[Conclusions] Fusion of background knowledge can supplement the semantic representation of short texts. Fusing user and content statistical features can assist semantic features in making decisions and improve the model's accuracy.