| 基于GA与SA的社区检测优化算法研究 |
| |
| 引用本文： | 王 妍,吴克晴,刘松华.基于GA与SA的社区检测优化算法研究[J].教育技术导刊,2019,18(9):77-80. |
| |
| 作者姓名： | 王 妍 吴克晴 刘松华 |
| |
| 作者单位： | 江西理工大学 理学院，江西 赣州 341000 |
| |
| 基金项目： | 国家自然科学基金项目（61762047）；国家重点研发计划项目（2016YFB0800700）；江西省科技厅青年科学基金项目（20161BAB211015） |
| |
| 摘 要： | 社区结构是网络最重要的属性之一，近年来社区检测受到极大关注，出现了很多社区发现算法。模块度是衡量社区划分好坏的重要指标，但是其分辨率却有一定局限性。将模块度中加入一个可调参数，根据社区结构调整参数更适合于需求不同的社区检测。随着网络规模的扩大，社区发现算法既要有较高的准确性，又要有很低的时间复杂性。提出一种发现算法GASA，该算法将遗传变异与模拟退火相结合，既有遗传算法的全局搜索能力，又有模拟退火算法的局部搜索能力。该算法用于社区检测优势明显，检测到的社区更接近真实社区。
|
| 关 键 词： | 模块度 遗传变异算法 模拟退火算法 社区检测 |
| 收稿时间： | 2018-12-29 |
A Community Detection Optimization Algorithm Based on GA and SA |
| |
| WANG Yan,WU Ke-qing,LIU Song-hua.A Community Detection Optimization Algorithm Based on GA and SA[J].Introduction of Educational Technology,2019,18(9):77-80. |
| |
| Authors: | WANG Yan WU Ke-qing LIU Song-hua |
| |
| Institution: | College of Science， Jiangxi University of Science and Technology， Ganzhou 341000， China |
| |
| Abstract: | Community structure is one of the most important attributes in the network. In recent years， community detection has attracted great attention， and many community discovery algorithms have emerged. Modularity is an important index to measure the quality of community division， but its resolution has some limitations. This paper adds an adjustable parameter to modularity， which can be adjusted according to community structure， and is more suitable for community detection with different needs. With the enlargement of network scale， community discovery algorithm not only needs to satisfy higher accuracy， but also reduces the computing time of the algorithm. So a discovery algorithm GASA is proposed， which combines genetic mutation with simulated annealing. It has both global search ability of genetic algorithm and local search ability of simulated annealing algorithm. This algorithm has stronger advantages in community detection， and the detected community is closer to the real community. |
| |
| Keywords: | cluster modularity GA SA community detection |
|
| 点击此处可从《教育技术导刊》浏览原始摘要信息 |
| 点击此处可从《教育技术导刊》下载免费的PDF全文 |
