基于构造型神经网络的分类算法

来源期刊:中南大学学报(自然科学版)2009年第3期

论文作者:刘承水

文章页码:737 - 741

关键词:模式识别;神经网络;覆盖;神经元;分类算法

Key words:pattern recognition; neural networks; covering; neuron; classification algorithm

摘    要:提出一种基于构造型神经网络的最大密度覆盖分类算法,以便更加有效地解决模式识别的问题。首先,引入一个密度估计函数,用该函数对样本数据进行聚类分析,找出同类样本中具有最大密度的样本数据点,然后,在特征空间里作超平面与球面相交,得到1个球面覆盖领域,从而将神经网络训练问题转化为点集覆盖问题。该算法的特点是直接对样本数据进行处理,有效地克服了传统神经网络训练时间长、学习复杂的问题,同时也考虑了神经网络规模的优化问题。计算机仿真实验结果证实了该算法的有效性。

Abstract: A new maximum density covering classification algorithm based on constructive neural networks was proposed, which can be used to resolve the problem of pattern recognition more effectively. Firstly, a density estimating function was proposed, which was used for clustering analysis of sample data, and a sample data point with the maximum density was found. Then, a super-plane was made to intersect a sphere in the characteristics of the space, and a spherical covering area was obtained, by which the training problem of neural networks can be transformed into the covering problem of a point set. The characteristic of the algorithm is that the sample data can be handled directly. This new algorithm can reduce the long training time and learning complexity of traditional neural networks. The optimization of the neural network is also considered. The simulation results show that the proposed neural network is quite efficient.

基金信息:国家自然科学基金资助项目

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号