摘要
BP人工神经网络算法作为一种基本算法在许多领域中都有着广阔的应用前景.但由于其本身的缺陷而很难投入实际应用.给出了一种基于SVM(Support Vector Machines)的BP改进回归算法,有效地提高了算法的收敛速度,减少了迭代次数,提高了回归预测的精度.
As a basic algorithm of artificial neural network, Back-Propagation（BP） network can be used in many fields. However, it is very difficult to use in real systems because of its shortcoming. In this paper, we propose a kind of regressive BP algorithm based on Support Vector Machines（SVM）. The SVM selects some optional vectors in training vector sets. It optimizes inputs of the BP network, Therefore, this algorithm increase the convergence speed of BP algorithm effectively, reduces the iterations greatly and improves regressive precision of prediction perfectly.
出处
《湖南工程学院学报（自然科学版）》
2007年第1期52-54,共3页
Journal of Hunan Institute of Engineering(Natural Science Edition)
关键词
神经网络
支持向量机
BP学习算法
回归
neural network
Support Vector Machines
BP algorithm
regressive algorithm
