Social network rumor recognition method and system

Abstract

The invention discloses a social network rumor recognition method and system. The method comprises the steps of: obtaining a microblog information case, obtaining microblog information and user information of the microblog information case, and according to the microblog information and the user information, extracting microblog content features of the microblog information case, wherein the microblog content features include a shallow text feature and a deeply implied microblog feature; extracting basic attribute features of a user and deeply implied features of the user according to the user information, and extracting microblog popularity features according to the microblog information, wherein the microblog popularity features include a volatility feature based on popularity and popularity trend, a difference feature and a forwarding feature; and establishing a feature vector and a training classifier according to the shallow text feature, the deeply implied microblog feature, the basic attribute features, the deeply implied features of the user and the microblog popularity features, inputting the feature vector into the classifier, and outputting a result.

Description

technical field [0001] The invention relates to the field of social network analysis, in particular to a method and system for identifying social network rumors. Background technique [0002] The popularity and popularization of social networks has led to an explosive growth in the amount of information in social networks, but the quality of information has not been improved accordingly. All kinds of spam information, especially false information such as rumors, is flooding the entire social network. The spread of rumors and Diffusion has brought great harm and negative impact to people's life and social development. [0003] Being able to identify rumors in social networks in a timely and accurate manner will not only help create a good Internet environment, help people better identify the authenticity of information, and prevent serious harm caused by malicious rumors in a timely manner. Play an active role in information guidance and other aspects. [0004] The existing...

AI Technical Summary

Problems solved by technology

[0004] The existing rumor identification methods can be mainly divided into two categories. One is based on artificial methods. Its mechanism is mainly through manual reporting and judgment of published news. Such methods cannot curb the spread of rumors in the early stages of generation. The timeliness is poor, and it requires a lot of labor and financial resources, and the cost is high; the other method is based on machine learning, which treats whether Weibo is a rumor as a classification problem, and uses various characteristics of Weibo, A certain classification learning algorithm is used to identify rumors. In terms of the selection of classification features, it can be divided into three types at present, namely, the content of Weibo, the publisher, and the dissemination of Weibo. In the selection of content features, currently the main It uses the superficial text features of the content (such as whether the content contains links, pictures, whether others are mentioned, etc.), without doing a deeper analysis of the text to fully mine its semantics, themes, emotions and other hidden features; In the aspect of microblogging, it mainly selects some static features, including basic attributes such as the number of followers and friends of the publisher, without considering the credibility and influence of the publisher, etc. In the selection of microblog communication features, related work mainly focuses on Concentrate on researching the propagation model of Weibo rumors, build a forwarding relationship graph with rumors as the original node, and simulate its propagation behavior, or only limit to some simple forwarding attributes, without further in-depth analysis of other characteristics of rumors in the propagation process

In these researches on feature selection for rumor recognition, the selected features are not well graded, and there are certain limitations, resulting in a poor final rumor recognition effect. To sum up, the existing methods lack a method that can accurately identify An automated approach to Weibo rumors

Images