基于亲和传播的动态社会网络影响力扩散模型

doi:10.11959/j.issn.1000-436x.2016194

通信学报 ›› 2016, Vol. 37 ›› Issue (10): 40-47.doi: 10.11959/j.issn.1000-436x.2016194

基于亲和传播的动态社会网络影响力扩散模型

陈云芳¹,夏涛^1,²,张伟¹,李晋³

¹ 南京邮电大学计算机学院，江苏南京 210003
² 中国电信济宁分公司，山东济宁272000
³ 北京信息科技大学公共管理与传媒学院，北京 100192

出版日期:2016-10-25 发布日期:2016-10-25
基金资助:
国家自然科学基金资助项目;北京市教育委员会人文社会科学研究计划面上基金资助项目

Influence diffusion model based on affinity of dynamic social network

Yun-fang CHEN¹,Tao XIA^1,²,Wei ZHANG¹,Jin LI³

¹ School of Computer Science and Technology,Nanjing University of Posts and Telecommunications,Nanjing 210003,China
² Jining Branch of China Telecom.,Jining 272000,China
³ School of Public Administration and Communication,Beijing University of Information Science and Technology,Beijing 100192,China

Online:2016-10-25 Published:2016-10-25
Supported by:
The National Natural Science Foundation of China;Humanistic and Social Science Research Plan Project of Beijing Municipal Education Commission

摘要/Abstract

摘要：

影响力最大化模型研究是近来社会网络的一个热点问题，然而传统的独立级联模型以静态网络中为基础，且激活概率一般设定为固定值。提出一种加入衰减因数的动态社会网络影响力扩散模型—DDIC 模型，其采用亲和传播来计算节点之间的激活概率，依据时间片对社会网络进行动态切分，使激活概率在不同时间片中实现了有效关联。实验结果表明DDIC模型中种子节点有更多机会激活它的邻居节点，且采用亲和传播计算出的影响力值能更准确地体现DDIC模型的传播过程。

关键词: 动态社会网络, 影响力扩散, 亲和传播

Abstract:

Recently,influence maximization model is a hot issue in the field of social network influence,while the traditional independent cascade model is generally based on static network with a fixed value of activation probability.DDIC model,which was a dynamic network influence diffusion model with attenuation factor was proposed.It calculated the activation probability between nodes via affinity propagation,and according with dynamic segmentation of social network time slice,calculation of influence on proliferation of next time slice with the current time slice of activation probability performance decay.The experimental results show that the nodes in the DDIC model have more chances to active the neighbor and the average probability of activing of the DDIC model is higher.Further experiments show that influence value via computing with affinity propagation can reflect the process of the spread model more accurately.

Key words: dynamic social network, influence diffusion, affinity propagation

陈云芳,夏涛,张伟,李晋. 基于亲和传播的动态社会网络影响力扩散模型[J]. 通信学报, 2016, 37(10): 40-47.

Yun-fang CHEN,Tao XIA,Wei ZHANG,Jin LI. Influence diffusion model based on affinity of dynamic social network[J]. Journal on Communications, 2016, 37(10): 40-47.

图/表 9

图1

图2

图3

表1

图4

图5

表2

图6

表3

参考文献 11

[1]	RICHARDSON M , DOMINGOS P . Mining knowledge-sharing sites for viral marketing[C]// KDD '02 Proceedings of the Eighth ACM SIGKDD International Conference. 2002: 61-70.
[2]	KEMPE D , KLEINBERG J M , TARDOS A . Maximizing the spread of influence through a social network[C]// The 9th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 2003: 137-146.
[3]	LESKOVEC J , KRAUSE A , GUESTRIN C . Cost-effective outbreak detection in networks[C]// The 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2007: 420-429.
[4]	CHEN W , WANG Y , YANG S . Efficient influence maximization in social networks[C]// Proceedings of the 15th ACM SIGKDD International Conference. 2009: 199-208.
[5]	GALSTYAN A , MUSOYAN V , COHEN P R . Maximizing influence propagation in networks with community structure[J]. Physical Review E Statistical Nonlinear ＆ Soft Matter Physics, 2009,79(2): 711-715.
[6]	CAO T Y , WU X D , WANG S ,et al. OASNET:an optimal allocation approach to influence maximization in modular social networks[J]. ACM Symposium on Applied Computing, 2010: 1088-1094.
[7]	SAITO K , NAKANO R , KIMURA M . Prediction of information diffusion probabilities for independent cascade model[J]. KnowledgeBased Intelligent Information and Engineering Systems Lecture Notes in Computer Science, 2008: 67-75.
[8]	BARBIERI N , BONCHI F , MANCO G . Topic-aware social influence propagation models[J]. Knowledge and Information Systems,2013: 2012,37(3): 81-90.
[9]	TANG J , SUN J M , WANG C . Social influence analysis in large-scale networks[C]// The 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2009: 807-816.
[10]	FREY B J , DUECK D . Clustering by passing messages between data points[J]. Science, 2007,315(5814): 972-976.
[11]	HABIBA T W B , BERGER-WOLF T Y . Maximizing the extent of spread in a dynamic network[J]. Technical Report 20.DIMACS, 2007.

时间序列	边数
T₁	631
T₂	617
T₃	627
T₄	625
T₅	627
T₆	629
T₇	603
T₈	622
T₉	632
T₁₀	662

时间序列	亲和传播平均概率
T₁	0.039
T₂	0.040
T₃	0.039
T₄	0.039
T₅	0.040
T₆	0.038
T₇	0.040
T₈	0.038
T₉	0.038
T₁₀	0.039

方法									排名前20的影响力节点
平均概率为0.04	1	5	0	3	2	13	23	41	20	61	14	7	52	18	19	4	24	49	82	158
亲和传播概率	1	5	0	2	13	3	12	41	19	92	61	21	52	18	158	15	120	38	49	103

基于亲和传播的动态社会网络影响力扩散模型

Influence diffusion model based on affinity of dynamic social network

在线阅读

PDF下载

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 11

相关文章 0

Metrics

推荐阅读 0