基于量子蚁群改进的K-means算法

Improved K-means Algorithm Based on Quantum ant Colony Algorithm

传统的K-means算法局部搜索能力强,但是对初始化比较敏感,并且容易陷入局部最优值,这些缺陷严重限制了它的应用范围;针对目前普遍所存在的问题,本文提出一种改进的基于量子蚁群的聚类方法,将量子计算原理和蚂蚁算法结合来改进K-means算法,该方法结合了两个方法的优点,力求优势互补,并且在该方法中引入微观适应性策略改进了算法中的交叉算子和变异算子,使得聚类算法的局部搜索能力得到很大的提高;实验证明该算法有很好的全局收敛性,克服了K-means的不足,能有效解决未成熟收敛的问题。

Traditional K means algorithm has big local search capability, but is more sensitive to the initialization, and is easy to fall in-to local optimum, these defects severely limits its scope of application. For the current widespread existence of the problem, this paper pres-ents an improved clustering method based on quantum ant colony. Quantum computing theory and ant algorithm to improve the K-means al-gorithm, this method combines the advantages of the two methods, and strive to complement each other. And the method to introduce micro adaptive strategy to improve the algorithm in the crossover operator and mutation operator to improve the local search ability of the clustering algorithm. Experiments show that the algorithm to ensure the diversity of the population, have a good global convergence, to overcome the deficiencies of the K means, can effectively solve the immature convergence.