政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/141762

政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/141762

English | 正體中文 | 简体中文 | Post-Print筆數 : 27 | 全文笔数/总笔数 : 113648/144635 (79%)
造访人次 : 51658746 在线人数 : 258

RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.

搜寻范围

查询小技巧：

您可在西文检索词汇前后加上"双引号"，以获取较精准的检索结果

若欲以作者姓名搜寻，建议至进阶搜寻限定作者字段，可获得较完整数据

进阶搜寻

主页 ‧ 登入 ‧ 上传 ‧ 说明 ‧ 关于政大典藏 ‧ 管理

到手机版

政大機構典藏 > 社會科學學院 > 財政學系 > 學位論文 > Item 140.119/141762

请使用永久网址来引用或连结此文件: https://nccur.lib.nccu.edu.tw/handle/140.119/141762

题名:	營業稅稽查關鍵因子：基於AI深度學習的探索 Detecting Business Tax Evasion: An AI Deep Learning Explorative Study
作者:	莊家明 Jhuang, Jia-Ming
贡献者:	周德宇韓幸紋 Zhou, Te-Yu Han, Hsing-Wen 莊家明 Jhuang, Jia-Ming
关键词:	營業稅逃漏深度學習 PCA主成分分析 MDS降維隨機森林 Business tax evasion Deep learning PCA principal component analysis MDS dimensionality reduction Random forest
日期:	2022
上传时间:	2022-09-02 15:30:57 (UTC+8)
摘要:	租稅為國家重要財政收入來源，納稅義務人的逃漏稅行為不僅減損國家財政收入，更間接影響租稅的公平性及企業競爭。過往查稅人員多以經驗法則選案進行查核，在稽核效率上略顯不足，今年起政府已經引進AI查找營業稅逃漏，本研究為印證深度學習是否相對傳統的線性模型分析，更能夠準確預測廠商逃漏稅的情形，故使用深度學習中的隨機森林模型以及經由PCA主成分分析篩選過後的變數執行OLS線性分析，比較兩者的預測準確程度，亦利用MDS降維模型將變數分布降維至三維立體空間，觀察有無逃漏稅的廠商資料樣本的可視化分布情形，綜合以上方式找出營業稅逃漏的關鍵變數，確認深度學習的工具確實有助於查稅人員執行查找逃漏稅案件。本研究結果顯示，非線性分析相較於線性分析會更加準確，且透過視覺化的立體圖形也能引導查稅人員在找異常值時，能發現更多的資訊納入查找關鍵因子的考量。本篇研究作為AI深度學習應用於查找營業稅查漏上的初探，未來還有非常大的精進空間。 Tax is one of the most important source of national fiscal revenue. Tax evasion not only diminish national fiscal revenue, but also indirectly affect the fairness of taxation among tax payers and even interfere with industrial competition. In the past, tax inspectors followed rules of thumb to select cases to detect tax fraud with no clear guidance toward audit efficiency. Since 2021, the government has incorporated AI to detect business tax evasion. This study is to investigate whether deep learning is more accurate than traditional linear model analysis in the aforementioned tasks. To predict the tax evasion of in the manufacturing sector, the random forest model in deep learning and the variables filtered by PCA principal component analysis were used to perform OLS linear analysis to compare the prediction accuracy. The MDS dimensionality reduction model was also used to reduce the dimensionality of the variable distribution. By using the three-dimensional space, we present the visual distribution of the data samples of manufacturers with or without tax evasion labels. We that identify the key variables of business tax evasion based on the above methods, and confirm that the tools of deep learning indeed provide clear guidance for tax inspectors to search for tax evasion cases. The results of this study show that non-linear analysis is more accurate than linear analysis, and the visual three-dimensional graphics can also guide tax inspectors to develop more intuition when looking for outliers to be considered for tax fraudulent behavior . This research is by no mean a preliminary and explorative study on the application of AI deep learning to understand business tax evasion, and there are still more threads of research to be pursued in the future.
參考文獻:	王建得、封昌宏與黃福隆 (2009)，《二聯式收銀機發票逃漏稅態樣之探討》，財政部98 年度自行研究報告提要表，財政部臺灣省南區國稅局。全國法規資料庫，檢自 https://law.moj.gov.tw/。江美虹與吳朝欽 (2014)，「營業稅逃漏行為之研究－以留抵稅額為例」，《財稅研究》，44(1)，81-105。行政院主計總處，檢自 https://www.dgbas.gov.tw/mp.asp?mp=1。李永山與陳彥文 (2006)，「應用類神經網路於營業稅逃漏稅預測模式之建構」，《資管評論》，14，63-79。沈哲緯、蕭震洋、辜炳寰、曹鼎志與鄭錦桐 (2014)，「運用隨機森林演算法進行莫拉克颱風災區土石流發生因子關聯性分析」，《災害防救科技與管理學刊》，3(1)，41-67。周文賢 (2002)，「多變量統計分析」，台北：智勝文化。林惠英 (2006)，《應用資料探勘技術於營業稅逃漏稅選案之研究》，國立中正大學會計所碩士論文。唐銘恩 (2019)，《建立逃漏稅高風險行為防制規範及查核機制等問題探討（因應洗錢防制法新制措施）》，財政部 108 年度自行研究評獎績優報告輯要，財政部賦稅署。孫克難 (2014)，「財政收支、世代正義與稅制改革—臺灣經驗之探討(下)」，《財稅研究》，43(5)，1–19。翁至威 (2020)，「首次導入AI技術財政部訂智能稅務服務四年計畫」，經濟日報 (2020/10/12)，檢自 https://koin.kcg.gov.tw/?p=9013。翁至威 (2021)，「三大稅目導入AI選案查核防堵地下經濟」，經濟日報 (2021/09/21)，檢自 https://money.udn.com/money/story/6710/5760028。財政部網站，檢自 https://www.mof.gov.tw/htmlList/18。陳明進 (2006)，「稽徵機關稅務查核對營利事業短漏報所得之影響」，《經濟論文》，34(2), 213-250。陳俊哲 (2013)，「稅法不確定對營利事業短漏報所得之影響:台灣之實證」，《臺大管理論叢》，24(1)，285-319。陳時仲 (2015)，《隨機森林模型效力評估》，國立交通大學統計學研究所碩士論文。游敏惠與吳朝欽 (2014)，「營利事業所得稅逃漏之研究—以擴大書面審核制度為例」，《財稅研究》，44(3)，94–121。黃則強 (1999)，《營業稅逃漏﹕實務與模型分析》，國立政治大學財政研究所碩士論文。黃美祝、林世銘與黃玟心 (2012).，「前期選案查核經驗對後續年度營利事業租稅逃漏之影響」，《應用經濟論叢》，(92)，59-91。經濟部中小企業處，檢自 https://www.moeasmea.gov.tw/masterpage-tw。鄭伃君、楊子霆與韓幸紋 (2020)，「租稅稽查與廠商租稅逃漏-來自擴大書審廠商的證據」，《經濟論文》，48(4), 475–509。謝宜婷 (2022)，「台灣用AI提升事實查核效率」，新南向政策資訊平台 (2022/03/14)，檢自 https://reurl.cc/yrXZdq。 Aizenman, J. and Jinjarak, Y. (2008), “The collection efficiency of the Value Added Tax: Theory and international evidence,” Journal of International Trade and Economic Development, 17(3), 391-410. Almunia, M. and Lopez-Rodriguez, D. (2018), “Under the Radar: The Effects of Monitoring Firms on Tax Compliance,” American Economic Journal: Economic Policy, 10(1), 1-38. Asatryan, Z. and Gomtsyan, D. (2020), “The incidence of VAT evasion, “Leibniz Centre for European Economic Research, ZEW Discussion Papers, 20-027. Bergman, M., Nevarez A.(2006), “Do Audits Enhance Compliance? An Empirical Assessment of VAT Enforcement,“ National Tax Journal,(59),817-832. Fathi, B. and Esmaeilian M. (2012), “Evaluation of Value Added Tax (VAT) and Tax Evasion, “Current Research Journal of Economic Theory, 4(1), 1-5. Healy, P. M. and Palepu, K. G. (2003), “The Fall of Enron, “Journal of Economic Perspectives, 17(2), 3-26. Long, S. Schwartz R.(1987), “The Impact of IRS Audit on Taxpayer Compliance: A Field Experiment on Specific Deterrence, “Annual Meetings of the Law and Society Association Washington D.C. Luo, P., Song, D. and Chen, B. (2020), “Investment and financing for SMEs with bank- tax interaction and public-private partnerships, “International Review of Economics and Finance, 65, 163-172. Tait, M. A. A. (1988), “Value added tax: International practice and problems (Vol. 24) ,” International Monetary Fund.
描述:	碩士國立政治大學財政學系 109255024
資料來源:	http://thesis.lib.nccu.edu.tw/record/#G0109255024
数据类型:	thesis
DOI:	10.6814/NCCU202201416
显示于类别:	[財政學系] 學位論文

文件中的档案:

档案	描述	大小	格式	浏览次数
502401.pdf		2793Kb	Adobe PDF2	35	检视/开启

在政大典藏中所有的数据项都受到原著作权保护.

社群 sharing

著作權政策宣告 Copyright Announcement

1.本網站之數位內容為國立政治大學所收錄之機構典藏，無償提供學術研究與公眾教育等公益性使用，惟仍請適度，合理使用本網站之內容，以尊重著作權人之權益。商業上之利用，則請先取得著作權人之授權。
The digital content of this website is part of National Chengchi University Institutional Repository. It provides free access to academic research and public education for non-commercial use. Please utilize it in a proper and reasonable manner and respect the rights of copyright owners. For commercial use, please obtain authorization from the copyright owner in advance.

2.本網站之製作，已盡力防止侵害著作權人之權益，如仍發現本網站之數位內容有侵害著作權人權益情事者，請權利人通知本網站維護人員(nccur@nccu.edu.tw)，維護人員將立即採取移除該數位著作等補救措施。
NCCU Institutional Repository is made to protect the interests of copyright owners. If you believe that any material on the website infringes copyright, please contact our staff(nccur@nccu.edu.tw). We will remove the work from the repository and investigate your claim.

DSpace Software Copyright © 2002-2004 MIT & Hewlett-Packard / Enhanced by NTU Library IR team Copyright © - 回馈