Researchers, enterprises, and governments have made great efforts to detect misinformation promptly and accurately. Traditional solutions either examine complicated hand-crafted features or rely... Show moreResearchers, enterprises, and governments have made great efforts to detect misinformation promptly and accurately. Traditional solutions either examine complicated hand-crafted features or rely heavily on the constructed credibility networks to extract useful indicators for discerning false information. However, such approaches require insightful domain expert knowledge and intensive feature engineering that are often non-generalizable. Recent advances in deep learning techniques have spurred learning high-level representations from textual and image content and discovering diffusion patterns with various neural networks. Despite the progress made by these methods, they still face the problem of overdependence on the content features and fail to discriminate against the influence of each user involved in the process of rumor spreading. Different user-aspect information plays different roles in various stages of rumor diffusion, effectively extract features from each aspect, and aggregate the learned features into a unique representation, which has not been well investigated. To address these limitations, we propose a novel model, UMLARD (User-aspect Multi-view Learning with Attention for Rumor Detection), to effectively learn the representation of different views of the users who engaged in spreading the tweet, and fuse the learned features through the distinguishable fusion mechanism. Finally, we concatenate the learned user-aspect features with content features to form a unique representation and feed it into a fully connected layer to predict the label of rumors. Our experiments conducted on real-world datasets demonstrate that UMLARD significantly improves the rumor detection performance compared to state-of-the-art baselines. It also allows explainability of the model behavior and the predicted results. Show less
Information cascade size prediction is one of the primary challenges for understanding the diffusion of information. Traditional feature-based methods heavily rely on the quality of handcrafted... Show moreInformation cascade size prediction is one of the primary challenges for understanding the diffusion of information. Traditional feature-based methods heavily rely on the quality of handcrafted features, requiring extensive domain knowledge and hard to generalize to new domains. Recently, inspired by the success of deep learning in computer vision and natural language processing, researchers have developed neural network-based approaches for tackling this problem. However, existing deep learning-based methods either focused on modeling the temporal characteristics of cascades but ignored the structural information or failed to take the order-scale and position-scale into consideration in modeling structures of information propagation. This paper proposed a novel graph neural network-based model, called MUCas, to learn the latent representations of cascade graphs from a multi-scale perspective, which can make full use of the direction-scale, high-order-scale, position-scale, and dynamic-scale of cascades via a newly designed MUlti-scale Graph Capsule Network (MUG-Caps) and the influence-attention mechanism. Extensive experiments conducted on two real-world data sets demonstrate that our MUCas significantly outperforms the state-of-the-art approaches. Show less
Researchers have exerted tremendous effort in designing ways to detect and identify rumors automatically. Traditional approaches focus on feature engineering. They require lots of human actions and... Show moreResearchers have exerted tremendous effort in designing ways to detect and identify rumors automatically. Traditional approaches focus on feature engineering. They require lots of human actions and are difficult to generalize. Deep learning solutions come to help. However, they usually fail to capture the underlying structure of the rumor propagation and the influence of all participants involved in the spreading chain. In this study, we propose a novel participant level rumor detection framework. It explicitly models and integrates various fine-grained user representations (i.e., user influence, susceptibility, and temporal information) of all participants from the propagation threads via deep representation learning. Experiments conducted on real world datasets demonstrate a significant accuracy improvement of our approach. Theoretically, we contribute to the effective usage of data science and analytics for social information diffusion design, particularly rumor detection. Practically, our results can be used to improve the quality of rumor detection services for social platforms. Show less
Researchers have exerted tremendous effort in designing ways to detect and identify rumors automatically. Traditional approaches focus on feature engineering, which requires extensive manual... Show moreResearchers have exerted tremendous effort in designing ways to detect and identify rumors automatically. Traditional approaches focus on feature engineering, which requires extensive manual efforts and are difficult to generalize to different domains. Recently, deep learning solutions have emerged as the de facto methods which detect online rumors in an end-to-end manner. However, they still fail to fully capture the dissemination patterns of rumors. In this study, we propose a novel diffusion-based rumor detection model, called Macroscopic and Microscopic-aware Rumor Detection, to explore the full-scale diffusion patterns of information. It leverages graph neural networks to learn the macroscopic diffusion of rumor propagation and capture microscopic diffusion patterns using bidirectional recurrent neural networks while taking into account the user-time series. Moreover, it leverages knowledge distillation technique to create a more informative student model and further improve the model performance. Experiments conducted on two real-world data sets demonstrate that our method achieves significant accuracy improvements over the state-of-the-art baseline models on rumor detection. Show less