摘要
社区结构的挖掘问题已经成为复杂网络中重要的研究方向,其挖掘算法是关键的核心问题.为了提高对社区结构进行挖掘的准确度,提出一种基于差分演化思想的复杂网络社区挖掘算法(Differential Evolution Community Detection Algorithm,DECD).DECD算法设计了一种新的编码方式,以模块密度函数作为优化目标,通过差分演化算法对复杂网络实施有效划分.实验结果表明,新的编码方式提高了编码速度并解决了社区重复编码问题,同时DECD算法能够提高复杂网络中的社区结构挖掘的准确度.
Community detection has been an important research direction of the structure of complex network, whose mining algorithm is a crucial core issue. To improve the accuracy of community detection, a community detection algorithm based on the theory of differential evolution for complex network （Differential Evolution Community Detection Algorithm, DECD） is presented. In the study, the algorithm proposed a creative encoding mode and chose the modularity density function as the optimization objective for differential evolution algorithm to detect the structure of complex networks. Experimental results demonstrate that not only the encoding speed is optimized and the repetition encoding problems is solved by the creative encoding mode, but also the accuracy of community detection in complex networks is improved by DECD algorithm.
出处
《江西理工大学学报》
CAS
2016年第1期95-101,共7页
Journal of Jiangxi University of Science and Technology
基金
国家自然科学基金资助项目(61462036)
关键词
社会网络
社区挖掘
差分演化
编码方式
social network
community detection
differential evolution
encoding
