Link prediction algorithm based on local centrality of common neighbor nodes using multi-attribute ranking

Link prediction has become an important research topic in the field of complex networks. The purpose of link prediction is to find the missing links or predict the emergence of new links that do not present currently in a complex networks. Considering that the local centrality of common neighbor nodes have an important effect on the similarity-based algorithm, but every centrality measure has its own advantage and limitation. We proposed a multi-attribute ranking method based on the Technique for Order Preference by Similarity to Ideal Object (TOPSIS) to evaluate the local centrality of common neighbor nodes comprehensively. In order to make the local centrality indicator based on TOPSIS achieve better results, we also proposed a new weight calculation method for the attributes normalization matrix. Experimental studies on 6 real world networks from disparate fields verified the superiority of the algorithm proposed in this paper.