In the study of complex networks very often arises a problem of identifying the "most important" nodes. For this purpose, we have defined various centrality measures, and selecting the one that is best suited depends primarily on the type of complex network. Network key data about most important nodes 54 can accelerate dissemination of information in information networks, stop spreading social network epidemics, break a network of terrorists to prevent possible terrorist attacks, etc. Degree centrality is a measure defined as the number of edges that are incident with the node. For each directed network the in-degree counts the links pointing to the node, and the out-degree counts the links pointing from the node. Eigenvector centrality is assigned to the node in relation to the results of all nodes in the network. Katz’s centrality has the idea of adding a constant term to the measure of centrality and thus allowing a node with many neighbors to have greater value of centrality regardless of whether his neighbors have high centrality or not. The central idea of PageRank centrality is that the importance of the node relies on the importance of those nodes that point to that node. The idea of subgraph centrality is that the importance of the observed node can be evaluated in view of its belonging to the closed walks that begin and end in that node. Closeness centrality defines the average number of steps we need to reach one of the other nodes in the range from one network node. The betweenness centrality indicates how likely the node is located on the path between two nodes, and is also used as an indicator of where the network could collapse.