| Recently, the study of complex networks has received more and more attention in many scientific fields. Researchers coming from biology, physics, and computer are focus on revealing the inherent factors of network evolution. As an important branch of complex networks, link prediction can help us understand the mechanism of network evolution and deal with many issues within various disciplines.Link prediction aims to infer the existence of links between nodes, including prediction of existent yet unknown links and future links. Similarity-based link prediction algorithms have become hot topics in recent years. As a result, researchers have proposed a variety of methods based on local and global topological information. This paper mainly studies a link prediction algorithm based on label propagation, and focus on how to improve the accuracy of link prediction.Firstly, the background knowledge and basic content of link prediction in complex networks are described in detail in this paper, and according to the difference of network structure, we summarize similarity-based link prediction algorithms.Secondly, we present a simple but effective link prediction strategy based on label propagation, which mimics the communication between people naturally. At the same time, we use a novel idea to calculate the likelihood of possible links between arbitrary pair of nodes that exploits the labels in each other’s memory.Thirdly, we perform an experimental comparison of the proposed method against some classic similarity-based link prediction algorithms using real-world networks. The experimental results show that our method offers higher precision than these well-known approaches.Finally, we propose a novel insight for improving Louvain community detection algorithm by using our ILP measure. The experimental results show that the improved Louvain method can be more effective in detecting communities than the basic Louvain community detection algorithm. |
PDF Full Text Request