大数据思维在纳税评估选案建模中的应用

An Application of Large-scale Database on Simulating Tax Assessment Screening

纳税评估选案的实质是风险识别与控制,体现的是信息管税中的大数据思维。这就要求税务机关在建立健全涉税信息数据库的基础上,探索机器学习算法,有效识别并遴选出疑点纳税人。本文对大数据思维在纳税评估选案建模中的应用进行探索,通过建立相应的模型,着力展现BP神经网络模型在纳税评估选案中的理论适用性。

The essence of tax assessment screening is to identify and control tax risks, which reflects the idea of large-scale database on tax governance by information. Tax assessment screening based on big data requires the tax authorities establish the tax-related information database firstly, and then explores the machine learning algorithm to effectively identify the suspect taxpayers. This paper focuses on the exploration of big data thinking on simulating tax assessment screening. The paper builds up a BP artificial neural network model to verify the application of machine learning algorithm on tax assessment screening.