一种基于层次约简的多层网络社区发现算法

发布时间：2018-06-10 07:03

  本文选题：多层网络 + 社区发现　； 参考：《计算机与现代化》2017年06期

【摘要】：如何在多层网络中发现社区是一项巨大挑战。目前有些算法将多层网络表示成三阶张量,然后使用非负张量分解进行社区发现。但在多层网络的每层网络中存在很多社区之间的连接或每层网络都很稀疏的情况下,非负张量分解算法的准确率较差。为了解决这一问题,本文提出一种改进算法。先将原始多层网络进行层次约简,减少多层网络的层数,使其社区结构更加凸显,然后再使用非负张量分解算法进行社区发现。在人工数据集与真实数据集上的实验表明,本文所提出的框架在准确率上有明显的优势。
[Abstract]:How to find communities in multilayer networks is a huge challenge. At present, some algorithms represent multi-layer networks as third-order Zhang Liang, and then use non-negative Zhang Liang decomposition for community discovery. However, the accuracy of non-negative Zhang Liang decomposition algorithm is poor when there are many connections between communities in each layer of multilayer network or every layer network is sparse. In order to solve this problem, this paper proposes an improved algorithm. Firstly, the original multi-layer network is reduced to reduce the number of layers, so that the community structure is more prominent, and then the non-negative Zhang Liang decomposition algorithm is used for community discovery. Experiments on artificial data sets and real data sets show that the proposed framework has obvious advantages in accuracy.
【作者单位】： 北京交通大学计算机与信息技术学院交通数据分析与挖掘北京市重点实验室;
【基金】：国家自然科学基金资助项目(61403023) 教育部-中国移动科研基金资助项目(MCM20150513) 中国博士后科学基金资助项目(2015M580040)
【分类号】：TP301.6
，

本文编号：2002380