RESUMEN
Link prediction aims to learn meaningful features from networks to predict the possibility of topology. Most of the existing research on temporal link prediction is mainly aimed at networks with slow topology changes. They ignore the information of topology interval and link duration. This paper proposes a link prediction model named FastSTLSG. It can automatically analyze the features of the topology in a unified framework to effectively capture the spatio-temporal correlation of Mobile Ad Hoc Networks. First, we regard the changing topology as a chaotic system, transform it into a series of static snapshots based on the autocorrelation function; Next, the fast graph convolutional network efficiently analyses the topological relationships between nodes and reduces the computational complexity by importance sampling. Then, the gate recurrent unit captures the temporal correlation between snapshots. Finally, the fully connected layer reconstructs the topological structure. In addition, we take full advantage of least squares generative adversarial networks to further improve the performance of generator to obtain high-quality link prediction results. Extensive experiments on different datasets show that our FastSTLSG model obtains higher prediction accuracy compared with existing baseline models.
RESUMEN
The Delegated Proof of Stake (DPoS) consensus mechanism uses the power of stakeholders to not only vote in a fair and democratic way to solve a consensus problem, but also reduce resource waste to a certain extent. However, the fixed number of member nodes and single voting type will affect the security of the whole system. In order to reduce the negative impact of the above problems, a new consensus algorithm based on vague set and node impact factors is proposed. We first use fuzzy values to calculate the ratings of all nodes and initially determine the number of agent nodes according to the preset threshold value. Then, we judge whether a secondary screening is needed. If needed, calculating the nodes' impact factor based on their neighboring nodes, and combining their impact factors with adjacency votes to further distinguish the nodes with the same fuzzy value. In addition, we analyze the dynamic changes in the composition and scale of the agent node set and give its ideal size through testing. Finally, we compare the proposed algorithm with DPoS algorithm and existing fuzzy set-based algorithms in different scales and network structures. Results show that no matter in what kind of network structures, the effectiveness of the proposed algorithm is improved. Among which, the most noticeable improvement is seen in complex network structures.