一种基于信息熵的金融数据神经网络分类方法

讨论一种基于信息熵的神经网络数据分类方法,通过所有神经元的统计权重信息对输入数据进行投票分类.这种多层网络结构以及基于信息量的分割算法,使得它在数据分类问题上比现有的多数神经网络具有更好的表现.其并行的可扩展结构适合硬件实现,能够提高实际运算速度,适合用来处理金融方面高维度、复杂的海量数据问题.

A classification approach of neural networks based on entropy for financial data

Through the entropy estimation,information theory-based learning is performed locally at each neuron.The input data are classified by using the weighted statistical information from all the neurons.Classification method based on multi-layer structure and information results in a better performance in data classification than many other existing methods of neural networks.This architecture can extend to a large embedded system to handle complex financial problems.