?
On the Algorithms of Identifying Opinion Leaders in Social Networks
The present paper is dedicated to the algorithms of opinion leaders’ identification in social networks on the basis of social network theory and machine learning methods. The main applications of such algorithms for social network analysis are discussed, examples of existing opinion leaders search algorithms and their advantages and disadvantages are given. A modified algorithm for finding the most influential social network nodes, which takes into account not only the position of the node in the network, but also signs of its activity, such as the number of posts, the number of comments and the number of reposts is proposed. The suggested algorithm is more accurate than the algorithms, which are based only on the position of the node in the network. Limitations of using the algorithms as well as possible problems with machine learning application in the field of social network analysis are described.