基于改进量子粒子群和K-means聚类的葡萄客户分类方法
作者:
作者单位:

1.中国农业大学信息与电气工程学院;2.中国农业大学信息与电气工程学院,中国农业大学食品与安全北京实验室;3.中国农业大学工学院,中国农业大学食品与安全北京实验室

作者简介:

通讯作者:

中图分类号:

TP301.6

基金项目:

现代农业产业技术体系建设专项资金项目(CARS-29)


Grape Customer Classification Method Based on Improved Quantum Particle Swarm and K-means Clustering
Author:
Affiliation:

1.College of Information and Electrical Engineering, China Agricultural University;2.Beijing Laboratory of Food Quality and Safety, China Agricultural University;3.College of Engineering, China Agricultural University

Fund Project:

the Chinese Agricultural Research System (grant number CARS-29)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对传统K-means聚类算法受初始类中心影响导致聚类准确度较低的问题,利用量子粒子群优化算法全局搜索能力强、收敛速度快的优势,提出一种基于改进量子粒子群的K-means聚类算法。为防止量子粒子群优化算法陷入局部极值,采用具有高斯扰动的局部吸引子以提高种群跳出局部最优的能力;为提高算法的收敛速度,采用加权更新种群平均最优位置以充分发挥精英粒子的优势;通过对收缩-扩张因子和随机变量参数进行交叉实验,选出最佳参数组合策略。在标准测试函数上的仿真结果表明,本文的改进量子粒子群优化算法在寻优精度、收敛速度以及稳定性上都有显著提高;通过对比7种聚类算法在UCI数据集上聚类结果,本文提出的聚类算法具有更好的聚类性能,可以有效降低K-means对初始聚类中心的依赖。最后,将该方法应用于我国鲜食葡萄市场客户分类中,证明了该方法的有效性和实用性。通过实证分析,基于改进量子粒子群的K-means聚类算法结构简单、精度高,具有一定的推广性。

    Abstract:

    The original K-means clustering algorithm is seriously affected by initial centroids of clustering and easy to fall into local optima. To overcome these shortages, this paper used the quantum particle swarm optimization (QPSO) which had power ability of global search and quick convergence rate to optimize the initial clustering centers of original K-means algorithm. As the QPSO algorithm can easily fall into the local optimum, the local attractor with gauss disturbance was used to make the population jump out of the local extremum. To improve the convergence speed of the algorithm, the weighted average best position was used to take advantage of the elite particles. The contraction-expansion factors and random variables were combined in order to select the best parameter strategy. The simulation results on various benchmark problems showed that the optimization accuracy, convergence speed and stability of the improved optimization algorithm were significantly improved. Experimental results on the typical UCI datasets showed that the proposed method was superior to compared algorithms. Finally, this method was applied to the customer classification of table grapes, which proved the effectiveness and practicability of the proposed clustering algorithm. Through the empirical analysis, it is also proved that this model can be promoted and applied.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-09-17
  • 最后修改日期:2021-01-06
  • 录用日期:2021-01-19
  • 在线发布日期:
  • 出版日期: