基于粗糙集理论与支持向量机的纳税评估模型

A Tax Assessment Model Based on Rough Set Theory and SVM Algorithms

研究纳税评估是对纳税人纳税情况进行评价的一种管理模式。为了区分纳税人是否正常申报税款,融合了粗糙集理论与支持向量机机器学习方法的优点,提出了一种新的纳税评估模型。通过选取纳税申报表中的指标,建立纳税评估指标体系,并利用粗糙集理论对指标进行属性约简,采用支持向量机对纳税人进行分类处理,建立了纳税评估模型。最后对上述模型进行了实例验证,实验结果表明,模型具有良好的纳税评估预测性能。

Tax assessment is a new management pattern which assesses the tax payment of taxpayers. To classify whether the taxpayer declares normally or not, the SVM machine learning theory is combined with Rough Set theory to found a tax assessment model. Firstly index is selected from tax returns to build an index system of tax assessment, then the rough set theory is used to reduce the attributes and SVM is used to classify the taxpayers. Finally a tax assessment model based on rough set theory and SVM algorithms is proposed, which is applied to corporate income tax. Experimental results show that this model performs well both in data classification accuracy and predictive accuracy.