一种新的混合球壳形数据聚类方法

模糊C-球壳聚类算法(FCSS)广泛应用于模式识别与机器学习领域。传统FCSS算法采用梯度法和交替寻优策略来求解,因此对初始值比较敏感,往往只能得到模型的局部极值点,从而影响聚类的效果,甚至使所采用的方法失效。将粒子群优化算法(PSO)、遗传算法(GA)与传统FCSS算法相结合,得到一种有效的混合球壳聚类算法PSO-GA-FCSS,数值实验结果表明新算法的性能优于FCSS算法、遗传算法与FCSS相结合的算法GA-FCSS以及粒子群算法与FCSS相结合的算法PSO-FCSS。

A New Hybrid Approach for Spherical Shell Clustering Data

The fuzzy C-spherical shells(FCSS) algorithm is widely applied to pattern recognition and machine learning.The classical clustering algorithm is based on gradient method and alternative optimization strategy;its disadvantages are sensitive to the initial values and easy to trap into a local optimal solution,affecting the clustering effect,even causing a false result.An efficient hybrid spherical shell clustering algorithm called PSO-GA-FCSS is proposed in this paper,which is based on the combination of FCSS...