计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (17): 85-93.DOI: 10.3778/j.issn.1002-8331.1608-0324

• 大数据与云计算 • 上一篇    下一篇

基于联系度的主题关注网络社区发现方法研究

陈  晓1,2,3,郭景峰1,3,范超智1,3   

  1. 1.燕山大学 信息科学与工程学院,河北 秦皇岛 066004
    2.华北理工大学 迁安学院,河北 迁安 064400
    3.河北省虚拟技术与系统集成重点实验室,河北 秦皇岛 066004
  • 出版日期:2017-09-01 发布日期:2017-09-12

Community discovering based on connection degree in topic-attention network

CHEN Xiao1,2,3, GUO Jingfeng1,3, FAN Chaozhi1,3   

  1. 1.College of Information Science and Engineering, Yanshan University, Qinhuangdao, Hebei 066004, China
    2.College of Qian’an, North China University of Science and Technology, Qian’an, Hebei 064400, China
    3.The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province, Qinhuangdao, Hebei 066004, China
  • Online:2017-09-01 Published:2017-09-12

摘要: 目前,以兴趣或主题分享等为目的的兴趣型社交网络则引领着社交网络改革的浪潮。融合社交关系和兴趣爱好关系构建一个新型社交网络模型——主题关注模型。在此模型基础上,采用集对联系度刻画顶点间相似性度量指标,该度量方法可以更好地刻画网络结构特征,提高传统局部相似性度量指标对某些顶点间相似性值的计算精度,降低全局相似性度量指标的计算复杂度。综合考虑主题影响和社交关系,将集对联系度与凝聚型聚类算法相结合,提出一种新的主题社区发现方法。在Karate网络和豆瓣数据集上进行主题社区发现,实验结果表明,考虑主题影响的划分具有更好的社区结构。

关键词: 主题关注网络, 集对, 联系度, 社区发现

Abstract: At present, the interest’s social networks, which share interests or topics, lead to a reform wave of social network. A new social network model, the topic-attention model, fusing the social relations and the topic-attention relation is constructed. Based on this model, firstly, using set pair connection degree to define the similarity between vertices, the measurement can better describe network structure characteristics, overcome the under-estimating for some similarity between vertices based on traditional local structures, and reduce the computational complexity of global similarity indices. Secondly, considering the influence of the topic and the social relationships, a new method of community discovery is proposed based on the set pair connection degree and the clustering algorithm. Finally, in Karate network and douban data set, experiment on the topic community mining, the results show, considering the impact of the topic influence has a better community structure.

Key words: topic-attention network, set pair, connection degree, community discovering