摘要
将遗传算法优化误差反传(BP)神经网络的数据挖掘方法应用于纳税评估,为税务机关申报征收与税务稽查的数据挖掘建立分类预测模型.从分析的结果可以看出,模型具有良好的预测效果,能为提高税收管理的针对性和有效性提出一些参考建议,以提高税收征管的水平.
In this paper, the application of genetic algorithm optimized BP neural network in tax assessment is discussed. The optimal prediction model is established on the tax department＇s database of collection of taxes and inspection. Based on the mining result, model reflects good prediction effect; it is useful and helpful in levying and collection of taxes and in reinforcing management＇s pertinence and efficiency of tax revenue.
出处
《汕头大学学报（自然科学版）》
2008年第2期62-68,共7页
Journal of Shantou University：Natural Science Edition
基金
广东省自然基金资助项目(No:04010981)
关键词
数据挖掘
纳税评估
人工神经网络
BP神经网络
遗传算法
data mining
tax assessment
artificial neural network
back propagation neural network
genetic algorithm
