[1]王刚.一种融合贝叶斯概率的社区结构发现方法研究[J].计算机技术与发展,2019,29(01):110-113.[doi:10. 3969 / j. issn. 1673-629X. 2019. 01. 023]
　WANG Gang.Research on a Community Structure Detection Method Based onBayesian Probability[J].,2019,29(01):110-113.[doi:10. 3969 / j. issn. 1673-629X. 2019. 01. 023]
点击复制

一种融合贝叶斯概率的社区结构发现方法研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
29
期数:
2019年01期
页码:
110-113
栏目:
智能、算法、系统工程
出版日期:
2019-01-10

文章信息/Info

Title:
Research on a Community Structure Detection Method Based onBayesian Probability
文章编号:
1673-629X(2019)01-0110-04
作者:
王刚
安康学院,陕西 安康,725000
Author(s):
WANG Gang
Ankang University,Ankang 725000,China
关键词:
社区发现;  贝叶斯概率;  信息熵;  数据挖掘
Keywords:
community detection; Bayesian probability; information entropy; data mining
分类号:
TP301
DOI:
10. 3969 / j. issn. 1673-629X. 2019. 01. 023
文献标志码:
A
摘要:
社区结构通常具有动态、不对称、模糊的特性.为了更好地发现社区结构以及描述社区成员之间的关系,针对当前方法的一些不足,对利用贝叶斯概率来改进社区结构发现的方法进行研究.贝叶斯概率在描述成员之间动态、因果、模糊关系时具有优势,通过引入信息熵,提出了一种融合贝叶斯概率的社区发现方法.该方法首先计算成员之间的贝叶斯概率,研究贝叶斯关系网络构建方法,得出成员之间不对称贝叶斯概率矩阵;然后根据系统内信息的熵相对稳定的性质,把成员间贝叶斯概率作为信息熵的概率输入,计算出新成员加入后信息熵的变化值,根据熵值变化情况来确定成员是否属于社区,从而在发现社区结构的同时,也能描述社区成员之间的不对称、动态和模糊关系.实验结果证明了该方法的有效性.
Abstract:
Community structure is commonly dynamical,fuzzy and asymmetric. In order to find the community structure and describe the relationship between members,for the shortcomings of current methods,we try to study a new method of community detection based on Bayesian probability which has advantages in describing dynamic,causal and fuzzy relations among members. By introducing information entropy,we propose a community discovery method integrating Bayesian probability. It first calculates the Bayesian probability among members,studies the Bayesian network construction method,and obtains the asymmetric Bayesian probability matrix between members. Then according to the nature of relatively stable information entropy in the system,the Bayesian probability between members as probability input of information entropy,the variable of information entropy after new members joining is calculated to determine whether mem- bers belong to the community. Thus,the asymmetrical,dynamic and fuzzy relationships among community members can be described while the community structure is discovered. Experiment shows that the method is effective.

相似文献/References:

[1]张璋,张然,朱东生. 关系数据库中社区发现方法研究[J].计算机技术与发展,2014,24(08):108.
　ZHANG Zhang,ZHANG Ran,ZHU Dong-sheng. Research on Community Discovery Methods in Relational Database[J].,2014,24(01):108.
[2]顾炎,熊超. 动态社区的点增量发现算法[J].计算机技术与发展,2017,27(06):81.
　GU Yan,XIONG Chao. Vertex-based Incremental Algorithm for DynamicCommunities Detecting[J].,2017,27(01):81.
[3]张振宇,朱培栋,王可,等.拓扑结构与节点属性综合分析的社区发现算法[J].计算机技术与发展,2018,28(04):1.[doi:10.3969/ j. issn.1673-629X.2018.04.001]
　ZHANG Zhen-yu,ZHU Pei-dong,WANG Ke,et al.A Community Detecting Algorithm Based on Comprehensive Analysis of Network Topology and Node Attributes[J].,2018,28(01):1.[doi:10.3969/ j. issn.1673-629X.2018.04.001]
[4]王学军,李有红,李炽平.基于密度自适应聚类数的社区发现谱方法[J].计算机技术与发展,2019,29(05):81.[doi:10. 3969 / j. issn. 1673-629X. 2019. 05. 018]
　WANG Xue-jun,LI You-hong,LI Chi-ping.A Community Detection Spectral Clustering Method Based on Density Adaptive Generation Cluster Number[J].,2019,29(01):81.[doi:10. 3969 / j. issn. 1673-629X. 2019. 05. 018]

更新日期/Last Update: 2019-01-10