改进的遗传k-means算法及其应用

Improvement of k-means clustering algorithm based on genetic algorithm and its application

为降低k值的不确定性和初始聚类中心的随机性对聚类结果的影响,提出一种改进的遗传k-means聚类算法。采用并行计算的方式降低k值和初始聚类中心对聚类结果的影响,利用平均类内距和类间距设计适应度函数保证聚类结果的正确性,改进遗传算法的遗传算子来提高算法效率。通过UCI标准数据集验证了该算法的正确性和有效性,并应用于玉米良种选育中。实验结果表明,该算法能获得更优良的玉米品种,指导玉米选育工作。

To reduce influences of the uncertainty of k value and the randomness of the initial clustering center on clustering results,an improved k-means clustering algorithm based on genetic algorithm was proposed.The parallel computing was used to reduce the effects of k-value and initial clustering center on clustering results.The average intra-class distance and inter-class distance were used to design the fitness function to ensure the correctness of the clustering results,and the genetic operator of genetic algorithm was improved to improve the efficiency of the algorithm.The correctness and effectiveness of the algorithm are verified by the UCI standard dataset,and the algorithm is used to select maize varieties.The experimental results show that the algorithm can obtain better corn varieties and guide the corn breeding work.