| 摘要: |
| 针对RBF神经网络易于陷入局部最大值的缺点,把遗传算法引入RBF神经网络中,利用遗传算法具有全局搜索的优点,对RBF神经网络的权值进行优化,并把优化后的神经网络模型用于DNA序列的分类。仿真实验表明,采用遗传优化的RBF神经网络比传统RBF神经网络分类有更高的分类效率和正确率。 |
| 关键词: RBF神经网络 DNA序列分类 特征提取 遗传优化 |
| DOI： |
| 投稿时间：2013-02-15修订日期：2013-05-10 |
| 基金项目:广西教育厅科研项目(200911LX486)资助。 |
|
| GA-Based RBF Network Optimization |
|
YANG Jie
|
| (Department of Computer Science, Liuzhou Vocational Technological College, Liuzhou, Guangxi, 545006, China) |
| Abstract: |
| Because RBF neural network is easy to fall into the defects of local maxima, the genetic algorithm is introduced into the RBF neural network. The advantage of genetic algorithm on global search can optimize the RBF neural network weights and the optimized neural network model is further used to classify DNA sequences. Compared with traditional RBF neural network, the genetic optimized RBF neural network shows higher classification efficiency and accuracy. |
| Key words: RBF neural network DNA sequence classification feature extraction genetic optimization |
