To improve the ability to predict the evolution trend of public opinion, a public opinion evolution trend prediction model based on an improved wolf pack algorithm and optimized long-short term memory neural network is proposed. Use Halton Sequence to initialization to improve population diversity. Design step factor to perform Gauss-Sine perturbation transformation to improve wolf group exploration and development capabilities. Combine with the spiral in the whale optimization algorithm to improve the siege mechanism to enhance the local search ability of wolves. The bidirectional memory population is used to increase the cooperative ability of the wolf pack. The improved wolf pack algorithm (IWPA) is applied to the hyperparameter prediction of the LSTM neural network. Using keywords such as “COVID-19” and “Food Safety”, the experiment proves that the IWPA-LSTM neural network public opinion evolution prediction model has good accuracy and generality. The model is suitable for the prediction of various public opinion evolution trends.

Mission of this research is to better capture the complexity of large-scale financial systems and overcome the insufficiency of equilibrium based neoclassical finance and behavioral finance in revealing the mechanism of financial crisis and the emergence of order. This study started from the dilemma of existing theories and the motivation of introducting complexity science. Then advances of multi-agent simulation and complex network analysis are summarized and discussed as two fundimental instruments of complexity science. Research trends of complexity science in financial research, are proposed as the result of the discussions. This study provides methodology reference and analytical tools for the theoretical research and practical applications of complex financial systems in the new era.

According to the complex correlation characteristics of the actual interdependent network and the important evaluation indicators in the network, five different coupling methods are proposed, and nine interdependent network models are established. Considering the information transmission, redistribution and cascading failures in the network, a cascading failure model based on betweenness artificial flow models is established. We based on the game theory, attack-defense game problems of critical infrastructure are analyzed from the perspective of complex network, and the robustness of various interdependent networks is analyzed. We discovered the preferences of game participants in the interdependent networks, providing decision support for the protection of infrastructure networks.

Most current predation models describe the stochastic nature of the environment in terms of white noise, for the problem that the parameters in the model may satisfy the Ornstein-Uhlenbeck process in a real situation, a stochastic three-species predation model with refuge and Ornstein-Uhlenbeck process is proposed. The dynamic behavior of the model is analyzed by its formula, differential inequality and stochastic analysis theory. The existence and uniqueness of the global positive solution of the model are proved, sufficient conditions for the average persistence and extinction of each population were obtained separately, finally, python is used to carry out numerical simulation to verify the conclusions obtained in the theorem, and further study the impact of refuge on population size.

In order to better represent the rare entities with a small number of neighbors, this paper proposes a knowledge graph embedding model based on the nearest neighbors (NNKGE), which uses the K-Nearest Neighbor algorithm to obtain the nearest neighbors of the target entity as extended information. Based on this, the relational nearest neighbors-based knowledge graph embedding model (RNNKGE) is proposed. To generate an enhanced entity representation, the nearest neighbors of the target entity in relation are obtained by the improved K-Nearest Neighbor algorithm and encoded by the graph memory network. Through the analysis of the experimental results on the public datasets, the above two models outperform the benchmark model (CoNE) in the case of using only the nearest neighbor nodes, alleviating the data sparsity problem and improving the knowledge representation performance.

Rapid and correct identification of vulnerable lines is of great significance to maintaining the safety of the power grid and avoiding large-scale power outages. Based on this, this paper proposes an analysis scheme to establish a multi-link accident chain model based on new indicators and generate a chain fault diagram. The model is calculated and analyzed based on the complex network theory, and the line vulnerability is measured based on the system load loss rate. Finally, the IEEE39-bus system is taken as an example to verify the effectiveness of the method. The method breaks through the limitation of a single pair of single lines in the process of establishing an accident chain model by the traditional method and effectively realizes the identification of fragile lines, so as to maintain the safety of the power grid and improve the stability of the power grid.

In order to better understand the network, based on the similarity, let the nodes select multiple similar nodes to form similar node pairs. Through the Monte Carlo simulation results, a pairing algorithm based on the maximum node similarity and degree is proposed to discover the overlapping community structure of the network. Using multi-level most similarity to continue to optimize the community structure, find out the deep overlapping structure and sub-community structure of the network community. The proposed algorithm discovers the overlapping structure of the network based on the reason why the real network forms a community, and further optimizes the community structure, discovering the deep overlapping community structure of the network and its sub-community structure.

The advance-selling mode has prolonged the overall sales duration of products, which brings new problems to the decision-making of product ordering time for e-commerce platforms. In order to study and compare the product deposit, product price and e-commerce platform profit under the influence of two kinds of ordering time strategies, a two-stage leader-follower game model is established to analyze the influence of deposit sensitivity, balance sensitivity and manufacturing cost on ordering time selection and product deposit decision. The results show that the consumer’s sensitivity to the product deposit and the product balance will affect the wholesale price of the products in the delayed order model, while the wholesale price in the early order model is only affected by the sensitivity of the product deposit. Compared with the early order model, the delayed order model can be applied to consumers with more stringent product price sensitivity. The E-commerce platform’s advance-selling coupon/deposit strategy depends on the consumers’ sensitivity to the product balance.

In order to identify the important weighted motifs in the directed weighted networks, the directed weighted networks are transformed into label networks and the simple motifs are expanded to label motifs by defining the edge weights as strong and weak labels. For the label motifs of the three nodes, the time-consuming procedure of subgraph traversal is replaced by the estimated probability of the corresponding number of the motifs appear in the random networks, and the important label motifs in the directed weighted networks are identified by introducing a dynamic indicator associated with the label motif type. It is applied to the passing networks of Guangdong team and Liaoning team in the 2019—2020 finals of China Basketball Association (CBA). The important passing modes of the teams in the games and the important players in the corresponding modes are obtained. The important label motifs play a significant role in mining the important construction patterns and key nodes of the directed weighted networks.

In order to better reveal the propagation mechanism of urban traffic congestion, this paper proposes a two-layer network congestion propagation model coupled with multiple warning information subnetworks and traffic road subnetworks, and explores the propagation mechanism of urban road congestion risk under multiple warning information. The model establishes a state transfer tree based on propagation dynamics and analyzes the propagation threshold using microscopic Markov chain (MMCA). Finally, the impact of multiple warning messages on the propagation of urban traffic congestion is analyzed through simulation experiments. The experimental results show that promoting the propagation of "fast" warning information and inhibiting the spread of "short" warning information can play a positive role in reducing traffic congestion pressure.