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：
针对传统聚类算法聚类质量不够理想、自适应性不强和易陷入局部极小值等缺陷，提出一种基于改进免疫算法的数据聚类算法，该算法通过引入生物免疫系统中的精英保留策略和期望繁殖率，使适应度高的个体得到保留，浓度高的个体得到抑制，提高了算法的自适应性和精度，在后期通过利用混沌优化方法，使算法的局部搜索能力得到增强。实验结果表明，该算法比传统的聚类算法具有更好的性能。
英文摘要：
Aiming at the defect of traditional clustering algorithms for the poor quality of the clustering, week adaptivity and eas- ily trapping into local minima, an immune clustering algorithm based on the elitist strategy and chaos optimization is proposed. By introducing elitist strategy and expected rate of reproduction in the biological immune system, individual of high fitness is re tained and individual of high concentrations is inhibited. At the same time, the algorithm＇s self-adaptability and accuracy are im- proved. By taking advantage of the chaos optimization method in the late stage, the local search ability of the algorithm is strengthened. Experimental results show that the algorithm has better performance than the traditional clustering algorithm.
同期刊论文项目
同项目期刊论文
期刊信息
位置：