一种基于改进k-means的RBF神经网络学习方法

摘要

针对传统RBF神经网络学习算法构造的网络分类精度不高,传统的k-means算法对初始聚类中心的敏感,聚类结果随不同的初始输入而波动.为了解决以上问题,提出一种基于改进k-means的RBF神经网络学习算法.先用减聚类算法优化k-means算法,消除聚类的敏感性,再用优化后的k-means算法构造RBF神经网络.仿真结果表明了该学习算法的实用性和有效性.%Aiming at the low classification accuracy of network trained by traditional RBF neural networks learning algorithm, the traditional ￡-means algorithm has sensitivity to the initial clustering center. To solve these problems, an improved learning algorithm based on improved fc-means algorithm is proposed. The new algorithm optimizes &-means algorithm with subtractive clustering algorithm to eliminate the clustering sensitivity, and constructs RBF neural networks with the optimized ￡-means algorithm. The simulation results demonstrate the practicability and the effectiveness of the new algorithm.