基于BP神经网络的绍兴市财政收入预测
【摘要】 近半个世纪以来,随着我国经济的不断发展,以及科学技术的不断进步,我们国家对于人工神经网络的相应研发和发展也不断的深入和进步,同时在这一方面也取得了很大的进展和相应等成果,比如在人工智能方面以及各种信息处理,还有涉及到医学、经济学等诸多方面,都有很深层次的应用和处理。对于人工神经网络来说,它可以在用户提供了相应的样本之后,依照系统和相应的算法不断的修正其神经元的行为,来确定其连接的强度,在相应的基本框架结构确定之后,再根据相应的样本进行自然的改变和修休整。因此通常来说,所谓用户不需要再根据相应的样本去做算法的调整。因此在很多神经网络预测模式中,BP神经网络算法是最常用的方法之一。
本文建立了BP神经网络模型,通过绍兴市实际的财政收入为原始数据样本,对模型进行训练,然后对财政收入进行了预测,并计算出预测值和实际值的误差。通过一系列的验证发现该模型的预测还是较为准确的。
【关键词】 人工神经网络,BP神经网络,财政收入,预测
Shaoxing City's Fiscal Revenue Forecast Based on BP Neural Network
【Abstract】 In the past half century, with the continuous development of our economy, and the progress of science and technology, our country for the corresponding research and development and the development of artificial neural network has been deeply and progress, also has made great progress in this area and the corresponding results. For example, there are significant applications in pattern recognition, artificial intelligence, information processing, predictive estimation, medicine, and economics. the artificial neural network can accept the sample set submitted by the user and continuously modify the strength of the connection between the neurons used to determine the system behavior in accordance with the algorithm given by the system. And after the basic structure of the network is determined, this change is naturally performed according to the set of samples it accepts. In general, users do not need to make corresponding adjustments to the learning algorithm of the network according to the sample set they encounter. Among the many artificial neural network prediction models, BP neural network algorithm is one of the most commonly used methods.
This article established a BP neural network model. Through the actual financial income of Shaoxing City as the original data sample, the model was trained, then the financial income was predicted, and calculate the error between the predicted value and the actual value. Through a series of verifications, the prediction of this model is still relatively accurate.
【Key Words】 Artificial neural networks,BP neural network,Revenue,Prediction
表目录
表3.1绍兴市地方财政收入与各影响因子系数表
表3.2绍兴市地方财政收入各影响因素原始数值
表3.3财政收入预测值与实际值相对误差