Node Importance Evaluation Algorithm for Complex Network Based on Time Series and TOPSIS

Evaluation of the importance of nodes in complex networks has always been the focus of complex network node research, and has practical application value. At present, in the evaluation of node importance, most evaluation indicators and evaluation algorithms are applicable to static networks. For example, based on graph theory and statistical characteristics, node importance evaluation indicators, such as degree and eigenvector centrality, proposed from the perspective of nodes and paths, are applicable to some complex networks, and have certain applicability and limitations. Select multiple evaluation indicators, establish node evaluation index attribute sets, and use multi-attribute decision-making methods, such as principal component analysis, TOPSIS, and attribute reduction sets. In view of the fact that the complex network is constantly changing with time, this paper designs a comprehensive evaluation algorithm for nodes, node attributes and timing based on the TOPSIS algorithm. The Facebook data set for three consecutive months was used, divided into three time periods by month, and the algorithm was verified in chronological order and compared with the results of the TOPSIS algorithm and the PageRank algorithm. Experimental results show that the algorithm takes into account the importance of the nodes in each time period, and the evaluation results are reasonable, more in line with the actual dynamic changes of the nodes, and have higher accuracy.