Label Propagation Algorithm Based on 2-Hop Neighbor and Weighted Path for Community Detection

Community detection is a research focus in the field of complex networks and has wide applications in various domains. However, traditional label propagation algorithms and modularity maximization algorithms often fail to achieve optimal results in many community detection tasks. By observing the characteristics of real-world social networks, this paper proposes a label propagation algorithm based on 2-hop neighbor and weighted path (LPANW) for community detection. The proposed algorithm is compared with three other algorithms on both synthetic networks and real-world datasets, demonstrating better NMI index and robustness.