基于人工萤火虫的模糊聚类算法研究

摘要

Fuzzy C-means(FCM) clustering algorithm is one of the most commonly used methods in data mining, such as being sensitive to initial conditions, usually leading to local minimum results. Therefore, a new glowworm swarm optimization ( GSO) -based fuzzy algorithm ( GSFM) is put forward in this paper. GSFM algorithm uses the capacity of global search in GSO algorithm to seek optimal solution as initial clustering-centers of FCM algorithm, and then use FCM algorithm to optimize initial clustering-centers, so as to get the global optimum. Above all, it solves the problems of FCM. According to the test, compared with the FCM clustering algorithm, the new algorithm improves the optimization ability of the algorithm, the number of iterations is fewer, and the convergence speed is faster. In addition, there is also a large improved at the clustering result.%模糊C-均值(FCM)聚类算法是数据挖掘中常用的方法之一,但往往受到初始聚类中心影响,收敛结果易陷入局部极小值的问题.该文提出了一种基于人工萤火虫(GSO)的模糊聚类算法(GSFM).该算法引入了全局寻优能力强的人工萤火虫算法来求得最优解作为FCM算法的初始聚类中心,然后利用FCM算法优化初始聚类中心,最后求得全局最优解,从而有效克服了FCM算法的缺点.实验结果表明,新算法与FCM聚类算法相比,提高了算法的寻优能力,并且迭代次数更少,收敛速度更快,聚类效果更好.