欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
为提高聚类效果，提出了一种基于分组遗传算法的聚类新方法。以改进的分组编码方式表示种群中的个体并基于此制定了合理的种群初始化方案，采用改进的遗传操作算子和种群更新规则，利用遗传算法高效的全局搜索能力实现聚类。通过非线性排序选择机制和精英保留策略提高了遗传进化的稳定性；引入同类并行交叉和合并分割变异算子提高了算法运行效率，增强了全局寻优能力。实验结果表明，该聚类新算法能够自动获得最优聚类数和最优划分方案，具有良好的性能和聚类效果。
英文摘要：
In this paper, in order to improve the accuracy of clustering, a new clustering method based on grouping genetic algo- rithm is proposed. The algorithm represents individuals by improved grouping coding mode, and based on it a manner of initial popula- tion is formulated. The algorithm employs improved genetic operators and performs clustering with the effective global searching ability of genetic algorithm. The evolutionary stability of the algorithm is improved by applying nonlinear selection mechanism and elitism sche- ma. The operating efficiency and global search ability of the algorithm is improved by adopting parallel crossover and merging-splitting mutation. The experimental results indicate that the clustering method based on grouping genetic algorithm can automatically find the proper number of clusters and the proper partition from a given data set, and derive better performance and higher accuracy for cluste- ring problems.
同期刊论文项目
同项目期刊论文
A Fully Implicit Finite Difference Scheme for the Variable-Order Fractional Advection-Diffusion Equa
A New Method of Information System Processing Based on Combination of Rough Set Theory and Pansystem
Finite Difference approximations for the Two-side Space-time Fractional Advection-diffusion Equation
Monotone Iterative Technique of Periodic Solutions for Impulsive Evolution Equations in Banach Space
期刊信息
位置：