论文部分内容阅读
移动网络社区发现是面向特定移动用户推广移动网络服务的有效方法.利用实际的移动通信数据构建移动网络模型,提出基于用户网络影响的移动社区发现算法.该算法采用信号传递思想将移动用户对网络的影响转化为欧式空间向量关系,利用欧式距离计算用户相似度,基于高效的仿射传播聚类算法实现了移动通信网络的社区结构检测,同时标示了每个社区的核心用户.最后,通过实验验证了算法的有效性并分析了算法的相关参数选择.
Mobile network community discovery is an effective method to promote mobile network services for specific mobile users.Using actual mobile communication data to construct mobile network model and propose a mobile community discovery algorithm based on the influence of user network.This algorithm uses the idea of signal transmission to connect mobile users to the network , The Euclidean distance is transformed into the Euclidean space vector relationship, the Euclidean distance is used to calculate the user similarity, and the community structure of the mobile communication network is detected based on the efficient affine propagation clustering algorithm, and the core users of each community are marked.Finally, Verify the validity of the algorithm and analyze the choice of the relevant parameters of the algorithm.