税务风险行为造成财政收入严重损失，损害国家公共基础设施，扰乱公平竞争的市场经济秩序。近年来，受数据挖掘和人工智能等信息技术的推动，税务风险检测受到了广泛的关注。为促进税务风险检测方法的高质量发展，本文首次对全球现有的税务风险检测方法进行了全面的概述和总结。具体而言，首先讨论了税务风险行为的成因及其负面影响，以及税务风险检测的发展历程。随后，重点分析了全球范围内基于数据挖掘的税务风险检测方法。根据算法所采用的不同原理，现有的风险检测方法可分为基于关系的和非基于关系的两类，共14种风险检测方法，并对每种方法进行了深入的探讨和分析。最后，本文分析和讨论了当前数据驱动的税务风险检测方法面临的四个主要技术瓶颈，包括整合和利用财政和税收碎片化知识的困难、检测结果的不可解释性、风险检测算法的高成本以及现有算法对标记信息的依赖。通过对这些问题的研究，得出知识导向和数据驱动的大数据知识工程将是未来税务风险领域的发展趋势，即税务风险检测从信息化向智能化转型。

Tax risk behavior causes serious loss of fiscal revenue, damages the country’s public infrastructure, and disturbs the market economic order of fair competition. In recent years, tax risk detection, driven by information technology such as data mining and artificial intelligence, has received extensive attention. To promote the high-quality development of tax risk detection methods, this paper provides the first comprehensive overview and summary of existing tax risk detection methods worldwide. More specifically, it first discusses the causes and negative impacts of tax risk behaviors, along with the development of tax risk detection. It then focuses on data-mining-based tax risk detection methods utilized around the world. Based on the different principles employed by the algorithms, existing risk detection methods can be divided into two categories: relationship-based and non-relationship-based. A total of 14 risk detection methods are identified, and each method is thoroughly explored and analyzed. Finally, four major technical bottlenecks of current data-driven tax risk detection methods are analyzed and discussed, including the difficulty of integrating and using fiscal and tax fragmented knowledge, unexplainable risk detection results, the high cost of risk detection algorithms, and the reliance of existing algorithms on labeled information. After investigating these issues, it is concluded that knowledge-guided and data-driven big data knowledge engineering will be the development trend in the field of tax risk in the future; that is, the gradual transition of tax risk detection from informatization to intelligence is the future development direction.

・To the best of our knowledge, we are the first to systematically review the research progress and development trends of tax risk detection worldwide.

・We introduce the relevant background knowledge related to tax risk detection, including the causes and harms of tax risk behaviors, along with the development process of tax risk detection. In addition, we provide a formal definition of tax risk detection and introduce details of the input data.

・We comprehensively sort out the research on tax risk detection, divide the existing methods into two categories, then list and introduce the 14 kinds of methods identified. Furthermore, we summarize the advantages and disadvantages of each method.

・We summarize the main problems faced in current tax risk detection practice, and further suggest a list of future research direction.