The relevance of the issues of the analysis of scientific publications is due to the fact that with the of Internet technologies, it became possible to collect data on the publication citation network. Meanwhile, the current approach to the analysis of scientific publications is based on bibliometric indicators that take into account only the number of citations. However, network analysis, which is mainly used in the study of social networks, is becoming increasingly widely used. The author has developed a methodology that allows for an effective analysis of scientific publications based on network analysis methods alternative to bibliometric methods. As criteria for evaluating scientific publications based on network analysis, relevant measures of the centrality of the citation network nodes are established: centrality by degree of connectivity; centrality by proximity to other nodes; centrality by mediation; centrality by authority; centrality by concentration. The author presented the experiment result that allows validating the developed methodology of network analysis of the scientific publications significance. Scientometric databases were used as primary sources of data on publications, which make it possible to track the citation of publications and identify relevant citation networks. The application of the proposed network analysis methodology contributes to the identification of important publications in the development of the scientific direction.