other مقایسه درخت تصمیم(الگوریتم C5.0 و جنگل تصادفی) و ماشین بردار پشتیبان در اعتبارسنجی مؤدیان مالیاتی مدیریتی Management پژوهشي Research <span style="font-size:14pt"><span style="line-height:115%"><span style="direction:rtl"><span style="unicode-bidi:embed"><span new="" roman="" style="font-family:" times=""><span lang="AR-SA" style="font-size:13.0pt"><span style="background:white"><span style="line-height:115%"><span b="" mitra="" style="font-family:"><span style="color:#181522">امروزه روش حسابرسی مبتنی بر ریسک در نظام&shy;های نوین مالیاتی مورد تأکید</span></span></span></span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:"> است، لذا تبیین مدلی جامع جهت رتبه&shy;بندی ریسک مؤدیان یکی از گام&shy;های اساسی اجرای طرح جامع مالیاتی است. بنابراین در این مقاله برآنیم تا عملکرد الگوریتم&shy;های درخت تصمیم و ماشین بردار پشتیبان را در </span></span></span><span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">اعتبارسنجی مؤدیان مالیاتی مقایسه کنیم. جامعه</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">آماری</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">این</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">پژوهش</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">شرکت&shy;های پذیرفته شده در بورس </span></span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">اوراق&shy;بهادار تهران هستند </span></span></span><span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">که طی سالهای 1392 تا 1397 فعال بوده&shy;اند و</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">برای</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">انتخاب</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">نمونه</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">از</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">روش</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">غربالگری</span></span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:"> (</span></span></span><span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">حذفی</span></span></span><span dir="LTR" style="font-size:13.0pt"><span style="line-height:115%"> (</span></span><span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">استفاده</span></span></span> <span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">شده&shy;است. </span></span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">در این پژوهش ابتدا با استفاده از <span style="color:black">تکنیک دلفی و متاسنتز </span>164 مؤلفه مؤثر در اعتبارسنجی مؤدیان مالیاتی شناسایی گردید. سپس داده&shy;های مورد نیاز برای اندازه&shy;گیری متغیرهای پژوهش از سایت کدال و با بررسی پرونده&shy;های مالیاتی استخراج شد و درنهایت با استفاده از داده&shy;های جمع&shy;آوری شده به بررسی میزان دقت </span></span></span><span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">درخت تصمیم (الگوریتم </span></span></span>&nbsp;<span dir="LTR" style="font-size:12.0pt"><span style="line-height:115%">C5.0</span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">و</span></span></span><b> </b><span lang="AR-SA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">جنگل تصادفی) و</span></span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:"> ماشین بردار پشتیبان</span></span></span><b> </b><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">در اعتبارسنجی مؤدیان مالیاتی </span></span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">پرداختیم. یافته&shy;ها نشان&shy;داد </span></span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">براساس </span></span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">نتایج حاصل از مقدار </span></span></span><a href="#_ftn1" name="_ftnref1" title=""><span class="MsoFootnoteReference" style="vertical-align:super"><span dir="LTR" style="font-size:12.0pt"><span style="line-height:115%"><span class="MsoFootnoteReference" style="vertical-align:super"><span style="font-size:12.0pt"><span style="line-height:107%"><span new="" roman="" style="font-family:" times="">[1]</span></span></span></span></span></span></span></a><span dir="LTR" style="font-size:12.0pt"><span style="line-height:115%">AUC</span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:"> الگوریتم</span></span></span><span dir="LTR" style="font-size:12.0pt"><span style="line-height:115%">C5.0 </span></span>&nbsp;<span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">و جنگل تصادفی برازش بهتری دارند با این وجود</span></span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:"> فرضیه پژوهش مبنی بر امکان پیش بینی ریسک مؤدیان مالیاتی با استفاده از الگوریتم </span></span></span><span dir="LTR" style="font-size:12.0pt"><span style="line-height:115%">SVM</span></span> <span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">رد نمی&shy;شود</span></span></span><span lang="FA" style="font-size:13.0pt"><span style="line-height:115%"><span b="" mitra="" style="font-family:">.<span style="color:red"></span></span></span></span></span></span></span></span></span> <div>&nbsp; <hr align="left" size="1" width="33%" > <div id="ftn1"><span style="font-size:10pt"><span style="font-family:Calibri,sans-serif"><a href="#_ftnref1" name="_ftn1" title=""><span class="MsoFootnoteReference" style="vertical-align:super"><span class="MsoFootnoteReference" style="vertical-align:super"><span style="font-size:10.0pt"><span style="line-height:107%"><span calibri="" style="font-family:">[1]</span></span></span></span></span></a><span arial="" dir="RTL" lang="AR-SA" style="font-family:">.</span> <span style="background:white"><span new="" roman="" style="font-family:" times=""><span style="color:#202124">Area under the ROC Curve</span></span></span></span></span></div> </div> <pre style="text-align:justify"> <span style="font-size:10pt"><span style="background:#f8f9fa"><span courier="" new="" style="font-family:"><span lang="EN" style="font-size:13.0pt"><span new="" roman="" style="font-family:" times=""><span style="color:#202124">Today, the risk-based audit method is emphasized in modern tax systems, so explaining a comprehensive model for rating the risk of taxpayers is one of the basic steps of implementing a comprehensive tax plan. Therefore, in this article, we aim to measure the performance of decision tree algorithms and support vector machine in the validation of taxpayers. The statistical population of this research is the companies accepted in the Tehran Stock Exchange, which were active during the years 2012-2017 and for the selection of the sample was made using the screening method (elimination).</span></span></span> <span lang="EN" style="font-size:13.0pt"><span new="" roman="" style="font-family:" times=""><span style="color:#202124">In this research, first, using Delphi technique and meta synthesis, 164 effective components in the validation of taxpayers were identified, then the data needed to measure the variables of the research were extracted from the Kodal website and by examining tax files, and finally by using the collected data, we investigated the accuracy of the decision tree (C5.0 algorithm and random forest) and support vector machine in validating taxpayers. The findings showed that based on the results of the AUC value, the C5.0 algorithm and the random forest have a better fit, however, the research hypothesis that it is possible to predict the risk of taxpayers using the SVM algorithm is not rejected.</span></span></span><span dir="RTL" lang="FA" style="font-size:13.0pt"><span new="" roman="" style="font-family:" times=""><span style="color:#202124"></span></span></span></span></span></span></pre> اعتبار سنجی مؤدیان, محافظه کاری, جریان وجه نقد عملیاتی, تمکین مالیاتی Taxpayer Validation, Conservatism, Operating Cash Flow, Tax Compliance 50 74 http://taxjournal.ir/browse.php?a_code=A-10-908-4&slc_lang=other&sid=1 Hoshang Khoramniya هوشنگ خرم نیا khoramniah@gmail.com 100319475328460010054 100319475328460010054 No Mirfeiz Fallahshams میرفیض فلاح شمس fallahshams@gmail.com 100319475328460010055 100319475328460010055 Yes Gholamreza Zomorodian غلامرضا زمردیان gh.zomorodian@gmail.com 100319475328460010056 100319475328460010056 No Ali Asghar Anvary Rostami علی اصغر انواری رستمی انواری رستمی anvary@modares.ac.ir 100319475328460010057 100319475328460010057 No