Compared with traditional Internet of Vehicles (IoV) in highways or suburban areas, maintaining network connectivity and stability is an important challenge for IoV in an urban scene, which has complex road conditions. Most existing studies focus on connectivity probability and whole network connectivity analysis. They analyze end-to-end connectivity probability under a certain node distribution model, and reveal the relationship between whole network connectivity and node density. Most of their results are not applicable to IoV in an urban scene. This work presents a concept of an IoV backbone net for IoV in an urban scene, called IoVB-net for short, to realize interconnections among nodes, and formulates an accessibility model based on IoVB-net, and proposes a routing method. The experimental results demonstrate that the trend of the accessibility's theoretical value is in accordance with that of data packet delivery rate, and the proposed routing method based on accessibility can also realize higher delivery rate and lower end-to-end latency than the improved greedy traffic aware routing and greedy perimeter stateless routing.

Graphs are widely used to model various practical applications. In recent years, graph convolution networks (GCNs) have attracted increasing attention due to the extension of convolution operation from traditional grid data to graph one. However, the representation ability of current GCNs is undoubtedly limited because existing work fails to consider feature interactions. Toward this end, we propose a Dual Feature Interaction-based GCN. Specifically, it models feature interaction in the aspects of 1) node features where we use Newton's identity to extract different-order cross features implicit in the original features and design an attention mechanism to fuse them; and 2) graph convolution where we capture the pairwise interactions among nodes in the neighborhood to expand a weighted sum operation. We evaluate the proposed model with graph data from different fields, and the experimental results on semi-supervised node classification and link prediction demonstrate the effectiveness of the proposed GCN. The data and source codes of this work are available at https://github.com/ZZY-GraphMiningLab/DFI-GCN .

Wireless cellular traffic prediction is a critical issue for researchers and practitioners in the 5G/B5G field. However, it is very challenging since the wireless cellular traffic usually show high nonlinearities and complex patterns. Most existing wireless cellular traffic prediction methods, lacking abilities of modeling the dynamic spatial-temporal correlations of wireless cellular traffic data, thus cannot yield satisfactory prediction results. To improve the accuracy of 5G/B5G cellular network traffic prediction, more cross-domain data was considered, a cross-service and regional fusion transfer learning strategy (Fusion-transfer) based on the spatial-temporal cross-domain neural network model (STC-N) was proposed. Multiple cross-domain datasets were integrated. The training accuracy of the target service domain based on the data characteristics of its source service domain according to the similarity between services and the similarity between different regions was improved, so the predictive performance of the model was enhanced. The experimental results show that the prediction accuracy of the traffic prediction model is significantly improved after the integration of multiple cross-domain datasets, the RMSE performance of SMS, Call and Internet service can be improved about 8.39%, 13.76% and 5.7% respectively. In addition, compared with the existing transfer strategy, the RMSE of the three services can be improved about 2.48%~13.19%.

Healthcare plays an increasingly essential role in our daily life. Modern Hospital Information Systems (HISs) record and store detailed medical treatment process information for all patients as event logs. By taking event logs as input, process mining techniques have been widely applied to extract valuable insights to improve medical treatment processes and deliver better healthcare services. However, considering the complexity of collaborations among different medical departments, existing model discovery techniques cannot be applied directly. To handle this limitation, this paper proposes a novel approach to support the discovery of Cross-department Collaborative Healthcare Process (CCHP) models from medical event logs. Specifically, an extension of classical Petri Nets with message and resource attributes is first introduced to formalize CCHPs. Then, a novel discovery algorithm is proposed to discover Intra-department Healthcare Process (IHP) models. Next, collaboration patterns among medical departments are formalized and corresponding discovery algorithms are given on that basis. Finally, a global CCHP model is obtained by integrating all discovered collaboration patterns and IHP models. By using four public medical event logs, we quantitatively compare our approach with the state-of-the-art process mining techniques in terms of model quality, and our experimental results demonstrate that the proposed approach can discover more accurate healthcare process models. Note to Practitioners—The recorded medical event logs by HISs can be used to extract valuable insights for the analysis of healthcare processes. However, existing process model discovery techniques cannot be applied for the analysis directly due to the complex collaborations among different medical departments of a hospital. This paper introduces a novel approach for cross-department collaborative healthcare process model discovery from medical event logs. All proposed techniques are fully implemented and publicly available. Using four public medical event logs, we show the applicability and advantages of our approach against existing ones. The proposed techniques are applicable to the model discovery and behavior understanding of real-life operational healthcare processes.