Network rumor detection method based on multi-modal relationship

Abstract

The invention discloses a network rumor detection method based on a multi-modal relationship. The network rumor detection method comprises the steps of obtaining a to-be-detected image and a related text published on a network platform; extracting visual feature vectors containing different types of objects in the image through a pre-training factor R-CNN model; after the text is preprocessed, performing semantic vector extraction through a GRU; capturing importance degrees of the visual feature vector and the semantic vector through an attention mechanism, and realizing cross-modal association between the image and the text so as to update the visual feature vector and the semantic vector; moreover, for the visual feature vector and the semantic vector, the relationship of internal dynamic information is modeled through an attention mechanism, so that the visual feature vector and the semantic vector are updated; and connecting the visual feature vector and the semantic vector obtained by updating the two parts together, and obtaining the probability that the information to be detected is the rumor and the real category through a binary classifier. The method can automatically judge whether the to-be-detected information belongs to the network rumors, and has relatively high detection accuracy.

Description

technical field [0001] The invention relates to the technical field of cyberspace security, in particular to a method for detecting network rumors based on multimodal relationships. Background technique [0002] The rise of the Internet society brings opportunities and challenges. In particular, the low barriers to access to the Internet and the freedom of information dissemination have seriously affected the stability of cyberspace. The wanton spread of Internet rumors is one of the problems that must be paid attention to. Today's social networking platform has already exceeded 100 million users, and its activity is extremely high. It has a wide spread, rapid spread, wide use, and is not limited by time and space. Its magnifying glass feature doubles the influence of information, especially some sensitive topics, focus Events, hot issues, major public events, and emergencies become widely known overnight, or cause a lack of trust, damage to the image of the government and e...

AI Technical Summary

Problems solved by technology

Although these methods provide heuristic ideas in the detection of disinformation in multimodal forms, they still have significant shortcomings.

First, the extraction process of image and text information is still relatively rough, especially the semantic features of pictures; second, in the feature fusion stage, the two modal features are often simply spliced ​​together, making it difficult to express the interaction and association between modalities

Images