Research On Network Reconstruction Based On Online Information System And Detection Of Fake News

More and more people use social platforms to obtain interesting information in recent years causing the boom of developing on social media.However,the information is mixed and equivocal,which leading to many problems in the governance of online information systems.For example,the spreading of untrue news makes a decline in social credibility,driving up the prices of goods to earn illegal profits and the issue of public opinion monitoring caused by the dissemination of inappropriate speech.This article cuts through the following two aspects to contribute to the governance of online information systems.One is from the dissemination process.When investigating the spreading of a piece of information or the diffusion of an innovation,we often lack information on the underlying propagation network.The important factor of spreading is that there are paths between the disseminators,and cutting off the process of spreading effectively plays a significant role in controlling the dissemination of information.So reconstructing the hidden propagation paths based on the observed diffusion process is a challenging problem which has recently attracted attention from diverse research fields.The other one is to detect the authenticity of news from the source of the dissemination.Misinformation has long been regarded as a severe social problem,where fake news is one of the most representative issues.What is worse,today's highly developed social media makes fake news widely spread at incredible speed,bringing in substantial harm to various aspects of human life.Yet,the popularity of social media also provides opportunities to better detect fake news.Making full use of existing information and realizing automatic detection of fake news has become another important issue in the current online information system.This paper proposes improved algorithms from these two aspects,which has important application value and social significance.The specific work is as follows:1.To address this reconstruction problem,based on static similarity metrics commonly used in the link prediction literature,we introduce new node-node temporal similarity metrics.The new metrics take as input the time-series of multiple independent spreading processes,based on the hypothesis that two nodes are more likely to be connected if they were often infected at similar points in time.This hypothesis is implemented by introducing a time-lag function which penalizes distant infection times.We find that the choice of this time-lag function strongly affects the metrics' reconstruction accuracy,depending on the network's clustering coefficient,and we provide an extensive comparative analysis of static and temporal similarity metrics for network reconstruction.Our findings shed new light on the notion of similarity between pairs of nodes in complex networks.2.Motivated by the above concerns,a novel detection framework,namely deep context attentional learning framework(DCAL)is proposed in this paper.Context information is crucial but not well studied in fake news detection.Thus,we model news context through iterative attention learning,and user context through hierarchical user attention,which can eliminate the influence of the water army through the user credibility.We conduct experiments on two real-world datasets,which demonstrate that the proposed joint model outperforms 6 state-of-the-art baseline methods for fake news detection(at least 2.9% improvement in Recall).Moreover,the proposed method is also explainable.