Information diffusion prediction is crucial for understanding social network dynamics, yet existing methods often neglect user participation uncertainty. This oversight typically stems from an implicit participation homogeneity assumption, which treats all observed interactions as equally reliable propagation signals, leading to fragile inferred topologies and uncertainty contamination. To address this, we propose SIEVE, a novel framework employing two synergistic strategies. First, robust node representations are learned via controllable uncertainty injection coupled with associated contrastive learning, mitigating topological fragility. Second, an uncertainty-aware directed graph aggregation mechanism is introduced, which dynamically constructs asymmetric aggregation topologies with adaptive weighting, thereby suppressing uncertainty contamination. Experiments on four public datasets demonstrate that SIEVE significantly outperforms state-of-the-art methods, offering valuable insights for designing robust information diffusion prediction models.

As a fundamental task in understanding the information diffusion process, information diffusion prediction has garnered significant attention in recent years. However, most existing studies tend to focus on the structural characteristics of information diffusion while neglecting an important phenomenon: the propagation of topics often carries multiple intentions, which align with specific user cognition at different evolutionary stages. This diversity in propagation intentions and the dynamic nature of user cognition pose challenges for prediction tasks. To address the above issues, this article introduces Buzz, an information diffusion prediction model, by innovatively approaching the problem from the perspective of aligning multiple propagation intentions with dynamic user cognition. First, to tackle the dynamic hierarchy of propagation intentions, a concise and efficient cascade intention extraction module is designed. This module uses observed diffusion cascades as intention anchors and employs an improved self-attention mechanism to generate representations of the current multiple propagation intentions. Based on this, attention weights are utilized to dynamically stratify the hierarchical structure of propagation intentions. Second, to address the dynamic nature of user cognition, we take social relationships as cognition anchors to initialize the latent diffusion network and dynamically weight the adjacency matrix through temporal slicing. This accurately models the dynamic diffusion process of topics. On this foundation, the cascades are segmented along the diffusion timeline into corresponding time slices, and graph convolution is applied to refine user dynamic cognition. Finally, considering the complexity of aligning propagation intentions with user cognition, we design a multihead attention fusion module. This module aligns propagation intentions based on user cognition, enabling more precise selection of target users. The great performance on four public datasets validates the effectiveness of our proposed approach.

The significant societal impact of online public opinion has spurred extensive research into the underlying mechanisms of information diffusion. While existing diffusion prediction models have effectively leveraged network structure, they often lack a deep understanding of the intrinsic patterns governing information dissemination, thus limiting their predictive power. To address this critical gap, we introduce PMRCA, a novel information diffusion prediction model inspired by social psychology and driven by fundamental propagation patterns. PMRCA posits that two key patterns underpin topic propagation: multisource resonance, where consensus emerges from consistent narratives across multiple sources, and cognitive adaptation, where information is tailored to align with audience cognitive features. Correspondingly, PMRCA incorporates two dedicated learning tasks: structural contrastive learning based on node homogeneity to capture multisource resonance, and clustering contrastive learning based on cognitive clusters to model cognitive adaptation. Extensive experiments on four real-world datasets demonstrate that PMRCA significantly outperforms existing mainstream models. Importantly, our pattern-driven approach is model-agnostic, allowing it to be flexibly applied to various diffusion prediction scenarios and enhance performance.

The rapid spread of online rumors has significant negative impacts on the online ecosystem and social order, which is closely tied to users’ cognitive traits. To explore the mechanisms of rumor propagation driven by cognition and mitigate its harm, we propose an information diffusion model based on the rumor, antirumor, and cognitive game. First, to address the uncertainty and difficulty in quantifying cognitive biases, and considering the advantages of fuzzy logic theory in handling uncertainty, a fuzzy logic-based algorithm for measuring user cognitive biases is proposed. Moreover, recognizing the nonlinear relationships among various factors, polynomial functions are introduced as the output of fuzzy rules to more accurately describe these complex relationships. Second, regarding the symbiotic and antagonistic relationships among multiple types of rumor information under the influence of cognitive biases, and leveraging the strengths of game theory in analyzing complex systems characterized by coexistence of symbiosis and antagonism, an evolutionary game-based rumor-antirumor user behavior mechanism is developed. This provides a robust theoretical foundation for understanding user state transitions and their evolutionary patterns. Finally, integrating the aforementioned research, and considering that the dynamic evolution of user cognition leads to variations in trust responses and attitudes toward rumor and antirumor information, the concepts of trust states—trust in rumor (TR) and trust in antirumor (TA)—are introduced into the classical susceptible-infectious-recovered (SIR) model. On this basis, a rumor propagation susceptible-trused-infectious-recovered (STIR) model incorporating user cognitive biases and evolutionary game theory is further constructed. Experimental results demonstrate that this model effectively reveals the game dynamics of multiple types of rumor information, providing a more efficient framework for studying rumor propagation in social networks.

The next point of interest (POI) recommendation task recommends POIs to users that they may be interested in next time based on their historical trajectories. This task holds value for both users and businesses. However, it has consistently faced the issue of cold-start caused by sparse user check-in data. Existing research mainly focuses on knowledge transfer among cities within the same data source, but these data are very rare. The abundance of available third-party data presents opportunities to improve cold-start performance, but it is not easy. This third-party data contain numerous entities, such as POIs and users, which have different representations and distributions across different data domains, making knowledge transfer difficult. To address these challenges, we propose the Tree-Based Cross-domain Knowledge Transfer (TCKT) model. First, we construct a multi-granularity Geographical Frequency Tree (GF-Tree), transforming the POI recommendation problem into a path generation problem. Second, we design a pre-training model to mine general user behavior patterns and spatio-temporal features among POIs from large-scale third-party data. Finally, we propose a dual-channel domain adaptation model to facilitate cross-domain knowledge transfer and improve cold-start performance. Experimental results on three public datasets demonstrate that our method outperforms state-of-the-art (SOTA) baseline methods.

This work is inspired by high-definition (HD) image generation techniques. When the user's interests are viewed as different frames of varying clarity, the unclear parts of one interest frame can be clarified by other interest frames. The user's overall HD interest portrait can be viewed as a fusion of multiple interest frames through detail compensation. Based on this inspiration, we propose a model for generating HD interest portrait called interest frame for recommendation (IF4Rec). First, we present a fine-grained pixel-level user interest mining method, Pixel embedding (PE) uses positional coding techniques to mine atomic-level interest pixel matrices in multiple dimensions, such as time, space, and frequency. Then, using an atomic-level interest pixel matrix, we propose Item2Frame to generate several interest frames for a user. The similarity score of each item is calculated to fill the multi-interest pixel clusters, through an improved self-attention mechanism. Finally, stimulated by HD image generation techniques, we initially present an interest frame noise compensation method. By utilizing the multihead attention mechanism, pixel-level optimization and noise complementation are performed between multi-interest frames, and an HD interest portrait is achieved. Experiments show that our model mines users' interests well. On five publicly available datasets, our model outperforms the baselines.

In online social networks, users’ awareness of a topic and their personal consciousness can significantly influence the dissemination of information on that topic. Traditional information dissemination research has mainly introduced user awareness as a single factor and has failed to fully consider the impact between users with different awareness. By considering the differences in user awareness and their impact, we proposed a multiple layer spatial information dissemination model based on user awareness. First, for the multidimensional nature of the communication space, we consider differences in users’ cognitive level into consideration, and by setting the corresponding threshold, the users who participate in topic dissemination are divided into two states of high awareness and low awareness. By constructing different communication networks, we depict the propagation dynamics of information at different levels of awareness and introduce an evolutionary game mechanism to analyze the influence between users at different levels of awareness networks. Second, during information dissemination, isolating transmission sources can provide effective control over the spread of information. Based on the susceptible-infectious-recovery (SIR) model in infectious disease dynamics, we consider that susceptible users have a certain chance of being isolated, so we abstract the isolated user group as an independent state Q (quarantine) and introduce it into the traditional SIR model to construct an SQIR information dissemination model. Finally, considering the influence between the information disseminated by high awareness users and low awareness users and the differences in users’ communication behavior at different periods, a dynamic information dissemination model based on user awareness and evolutionary game is obtained through a time-slicing process to dynamically explore the dynamics of information dissemination. The experiments showed that the proposed model could effectively reveal the influence of different levels of awareness on the dynamic information dissemination trend.