22 February 2023 in ACM Transactions on Knowledge Discovery From Data
With the wide use of Location-Based Social Networks (LBSNs), predicting user friendship from online social relations and offline trajectory data is of great value to improve the platform service quality and user satisfaction. Existing methods mainly focus on some hand-crafted features or graph embedding models based on the user-location bipartite graph, which cannot precisely capture the latent mobility similarity for the majority of users who have no explicit co-visit behaviors and also fail to balance the tradeoff between social features and mobility features for friendship prediction. In this regard, we propose a dual subgraph-based pairwise graph neural network (DSGNN) for friendship prediction in LBSNs, which extracts a pairwise social subgraph and a trajectory subgraph to model the social proximity and mobility similarity, respectively. Specifically, to overcome the co-visit data sparsity, we design an entropy-based random walk to construct a location graph that captures the high-level correlation between locations. Based on this, we characterize the pairwise mobility similarity from trajectory level instead of location level, which is modeled by a graph neural network (GNN) on a labeled trajectory subgraph composed of the two trajectories of the target user pair. Besides, we also utilize another GNN to extract social proximity based on social subgraph of the target user pair. Finally, we propose a gate layer to adaptively balance the fusion of the social and mobility features for friendship prediction. We conduct extensive experiments on the real-world datasets and demonstrate the superiority of our approach, which outperforms other state-of-the-art methods. In particular, the comparative experiments on the trajectory level mobility similarity further validate the effectiveness of the designed trajectory subgraph-based method, which can extract predictive mobility features.
26 November 2020 in Journal of Systems Science and Systems Engineering
26 February 2024 in IEEE Internet of Things Journal
Virtualized network function (VNF) and service function chain (SFC) are the fundamental components in network functions virtualization (NFV) infrastructure, which supports the evolution of modern 5G networks. For online Internet of Things (IoT) applications, characterized by dynamic and diverse requirements, achieving optimal quality of service hinges on a resource-efficient yet performant SFC caching strategy, which is a critical challenge. Besides, despite the performance boost and flexibility brought by modern cloud-native technology, it brings the cold-start problem due to the requirement for runtime image transmission and booting-up, resulting in a nonnegligible launch latency. To tackle these challenges, this article proposes cloud-native parallel SFC caching (CPSC) framework, a novel approach to address the CPSC problem in edge–cloud networks leveraging deep reinforcement learning (DRL), seeking an efficient resource utilization of the edge–cloud network with consideration of SFC processing performance and cold-start suppressing. Graph convolutional network (GCN)-based embeddings are adopted for topology-aware feature extraction of the substrate edge–cloud network as well as the incoming SFC caching requests. Then, a pointer network (PN) is utilized for contextual information-aware caching decision making. Benefiting from the online capability of DRL, CPSC makes caching decisions in an online manner with no prior knowledge requirement on future incoming requests. Extensive simulations show that CPSC manages to outperform the state-of-the-art approaches in edge network acceptance ratio and launch latency, with minimal overhead on the SFC processing performance and decision-making duration.
28 September 2023 in The Journal of Supercomputing
1 November 2020 in Chinese Journal of Electronics
Graph convolution networks are extremely efficient on the graph-structure data, which both consider the graph and feature information. Most existing models mainly focus on redefining the complicated network structure, while ignoring the negative impact of lowquality input data during the aggregation process. This paper utilizes the revised Laplacian matrix to improve the performance of the original model in the preprocessing stage. The comprehensive experimental results testify that our proposed model performs significantly better than other off-the-shelf models with a lower computational complexity, which gains relatively higher accuracy and stability.