A Novel Feature-based PageRank Algorithm for Node Importance Ranking

Node importance ranking is one of key problems in the study of complex networks. The classical PageRank algorithm only focus on the network structure, which lead to inaccurate ranking result. By introducing the features of node attributes and user preference, a novel feature-based PageRank (FBPR) algorithm is proposed to identify the important nodes accurately and efficiently. The weight matrix and the fixed teleportation vector are redesigned by the feature similarities in the FBPR model. For different application scenarios, we can get different ranking results by adjusting the node attributes factor and the user preference factor. Finally, several simulation experiments are presented to verify the effectiveness of the FBPR algorithm.