Predicting Viral News Events in Online Media

The information diffusion and dissemination define critical dynamics observed in large complex networks. The underlying information propagation topology, however, is often hidden or incomplete because of the lack of explicit citations of the sources. We proposed a scalable parallel algorithm to derive the node embeddings to better understand the information dissemination patterns and predict emergent cascades of viral events in online media. Unlike previous works which concentrate on modeling the links of information propagation, our algorithm infers the topic-specific output influence and the input selectivity of nodes. The parallel algorithm iteratively merges local node embeddings in particular communities to obtain the global optimal results so that the processing of cascades can be significantly accelerated. Based on the obtained latent representation of nodes, the emergent cascades of viral news events in online media can be successfully predicted with an 80\% accuracy at its early stage. Experimental results show that our parallel inference algorithm achieves a 10-fold acceleration and requires a low communication overhead, while the accuracy of the cascade size prediction is preserved.