论文部分内容阅读
为了解决动态虚拟企业中订单任务的分类问题,在分析SOFM网络学习算法的基础上,以提高算法的收敛速度和分类精度为出发点,研究了对SOFM网络的改进方法。针对SOFM网络中过多的输入不利于任务分类的问题,研究了基于粗糙集的订单任务属性特征提取方法。最后,在实际样本数据的基础上,以matlab为仿真工具,利用粗糙集的属性特征提取及改进的SOFM网络对样本进行分类,证实了该方法可以实现订单任务的自动分类,并且较传统SOFM网络分类方法具有更快的速度和更高的精度。
In order to solve the classification problem of order task in dynamic virtual enterprise, based on the analysis of SOFM network learning algorithm, the improved method of SOFM network is studied in order to improve the convergence speed and classification accuracy of the algorithm. Aiming at the problem that too much input in SOFM network is not conducive to task classification, this paper studies the feature extraction method of order task based on rough set theory. Finally, on the basis of the actual sample data, using matlab as the simulation tool, using the feature extraction of rough sets and the improved SOFM network to classify the samples, it is proved that the method can automatically classify the orders and tasks, and compared with the traditional SOFM network Classification method has faster speed and higher precision.