基于H指数的改进标签传播算法研究

发现复杂网络中的结构和特征是社区发现的一个重要任务.标签传播算法(LPA)因具有接近线性的时间复杂度,常用于快速处理大规模的社区网络.针对该算法在节点的更新顺序和标签选择策略上存在很大的随机性,严重破坏了算法的稳定性和社区划分结果的准确性.提出了一种基于节点H指数的标签传播算法,即利用节点的综合影响力改进标签传播算法的节点更新顺序和标签选择策略.实验研究表明,改进算法有效地降低了算法的随机性,提高了社区划分的稳定性和准确性.

An improved label propagation algorithm based on H-index

Discovering the structure and characteristics of complex networks is an important task for community discovery. The label propagation algorithm(LPA),which has a near-linear time complexity,is often used to quickly handle large-scale community networks. However,due to the randomness of the algorithm in the update order and label selection strategy,the stability of the algorithm and the accuracy of community partitioning results are seriously damaged. This paper proposes a label propagation algorithm based on the H-index,which uses the node influence to improve the node update order and label selection strategy of the label propagation algorithm. Experimental results show that the improved algorithm can effectively reduce the randomness of the algorithm and improve the stability and accuracy of community partition.