基于混合身份搜索黏菌优化的模糊C-均值聚类算法

针对模糊C-均值聚类算法(fuzzy C-means clustering, FCM)对于初始化聚类中心敏感、收敛速度慢，聚类效果不稳定且容易陷入局部最优等问题，提出了一种将黏菌(SMA)与青少年身份搜索(AISA)相融合的自适应优化模糊C-均值算法(AISA-SMA-FCM)。该算法首先通过引入AISA算法中的青少年社会机制，改善SMA算法中的全局搜索和局部开发性能。克服了SMA对于高维数据及部分混峰数据不敏感的缺陷，通过标准测试函数验证改进后的混合AISA-SMA算法寻优求解性能更为优秀；其次此算法用于FCM聚类算法的迭代机制中，通过将AISA-SMA聚类环节加入FCM算法聚类中心迭代过程中，使FCM算法获得自适应优化算法相同的特性，即算法在每次迭代中都将具有探索和开发两个过程，并依据循环迭代次数调节比重，求解聚类结果；最后通过UCI标准数据集仿真测试，利用适应度平均值与聚类正确率评价所提算法的稳定性与有效性，结果表明，AISA-SMA算法用于FCM聚类问题效果较好，AISA-SMA-FCM算法较其他聚类方式和相应的优化技术具有收敛速度快、求解精度高的优点。

An optimization fuzzy C-means clustering algorithm based on the hybrid identity search and slime mold algorithms

The fuzzy C-means clustering algorithm (FCM) has many shortcomings, such as high sensitivity to initial clustering centers, slow convergence, unstable clustering results, and ease of falling into local optimums. To address these problems, an adaptive optimization fuzzy C-means algorithm based on the fusion of the slime mold algorithm and the adolescent identity search algorithm (AISA–SMA–FCM) is proposed in this paper. First, the algorithm improves the global search and local development performance of SMA by introducing the youth social mechanism in AISA, thus overcoming the limitation of SMA of not being sensitive to high-dimensional data and some mixed peak data, along with an excellent performance of the improved AISA–SMA algorithm in optimizing and solving problems verified by the standard test function. Second, the novel proposed algorithm adds the AISA–SMA clustering link to the iteration process of FCM, enabling it to have the same characteristics as the adaptive optimization algorithm—undergoing two processes of exploration and development in each iteration, adjust the proportion as per the number of iterations, and solve the clustering results. Finally, through the simulation test on the UCI standard data sets, the stability and effectiveness of the algorithm are evaluated based on the fitness average and the clustering accuracy rate. The results show that the AISA–SMA algorithm demonstrates a good effect when used in the iteration mechanism of the FCM algorithm, with a faster convergence speed and a higher solution accuracy when compared with other clustering methods and corresponding optimization technologies.