Rumor detection method and system based on dynamic multi-hop graph attention network

Abstract

The invention relates to a rumor detection method and system based on a dynamic multi-hop graph attention network, and the method comprises the following steps: A, extracting the user information and reply information of posts under known events, and the forwarding or comment relation between the posts, marking whether each event is a rumor, and constructing a training set S; b, using the training set S to train a deep learning network model G based on a dynamic multi-hop graph attention network, wherein the deep learning network model G is used for analyzing whether each event is a rumor or not; and step C, inputting the extracted user information, reply information and post relationships and text information of the posts into a trained deep learning network model G to predict whether each event is a rumor or not. The step C is based on a rumor detection model of a dynamic multi-hop graph attention network, and the step C is that the extracted user information, reply information and post relationships and text information of the posts are input into a trained deep learning network model G to predict whether each event is a rumor or not. The method and the system are beneficial to improving the rumor detection accuracy.

Description

technical field [0001] The invention belongs to the field of natural language processing, and in particular relates to a rumor detection method and system based on a dynamic multi-hop graph attention network. Background technique [0002] With the development of the mobile Internet, social media has inevitably accelerated the spread of rumors while bringing us convenience in study and life. The proliferation of online rumors not only causes serious interference to people's production and life, but also destroys the social trust system to a certain extent, causing serious negative impacts on society. Although various media organizations have strengthened the monitoring of rumors and opened rumor-refuting platforms, these measures require a lot of manpower and material resources to collect information to verify the authenticity of the news, and there are problems such as time lag, incomplete information coverage, and event reversal. Therefore, it is necessary to automate the ...

AI Technical Summary

Problems solved by technology

In addition, forwarded posts occupy a large proportion in social media. Since forwarded posts themselves do not carry additional text information, existing research often chooses to ignore them. However, for forwarded posts, user information is an important link between forwarded posts and source posts. The difference, ignoring the forwarding of the post will lead to the inability to fully extract the user characteristics of the post, and will also cause the loss of key information such as the propagation path in the graph neural network

Images