本文選題：聚類　切入點：稅源專業(yè)化管理　出處：《集美大學》2017年碩士論文　論文類型：學位論文

【摘要】：繼云計算后,大數(shù)據(jù)(Big Data)成為信息技術(shù)領(lǐng)域最為熱門的概念之一。根據(jù)中國互聯(lián)網(wǎng)數(shù)據(jù)中心的報告,2012年全球的數(shù)據(jù)總量2.7ZB(1ZB=106PB,1PB=106GB),而加上互聯(lián)網(wǎng)和移動互聯(lián)網(wǎng),各類記載了個人購物、位置等方面的相關(guān)數(shù)據(jù)更是海量。所以如何利用這些數(shù)據(jù),成為了日益突出的問題。作為為國聚財、為民收稅的稅務(wù)部門,同樣存在著涉稅數(shù)據(jù)缺乏深度挖掘利用的問題。隨著“金稅三期”的上線,稅務(wù)部門的涉稅數(shù)據(jù)從省級集中變成了全國數(shù)據(jù)的大集中,稅務(wù)部門的數(shù)據(jù)庫中存在海量的外部和內(nèi)部數(shù)據(jù)。但是稅務(wù)部門對這些海量數(shù)據(jù)的分析與處理局限于匯總、查詢、對比等基礎(chǔ)性的應(yīng)用,無法對數(shù)據(jù)的深層信息進行挖掘。本文力圖在稅務(wù)部門的工作中引進數(shù)據(jù)挖掘技術(shù),從而讓稅務(wù)部門的決策更加具有科學性并提高納稅服務(wù)質(zhì)量,同時將研究的重心放到了聚類分析在稅源專業(yè)化管理的應(yīng)用上。稅源專業(yè)化管理是基礎(chǔ)的稅務(wù)管理工作,掌控全面的稅源信息,實行嚴謹?shù)膶I(yè)化管理方法,能夠充分保證稅務(wù)部門稅收的應(yīng)收盡收。聚類分析是數(shù)據(jù)挖掘技術(shù)的一種,目前已經(jīng)在商業(yè)、醫(yī)療等領(lǐng)域廣泛應(yīng)用。將聚類分析用于稅源專業(yè)化管理中,能夠充分地提升稅源分類的有效性,同時可以使稅務(wù)人員能夠科學地制定稅源管理策略,提高稅源專業(yè)化管理的效率,并有助于推動“放管服”改革不斷深入。本文在總結(jié)了國內(nèi)和國外研究與實踐的基礎(chǔ)上,梳理了稅源專業(yè)化管理、大數(shù)據(jù)、數(shù)據(jù)挖掘技術(shù)和聚類算法等方面的相關(guān)知識,然后把各方面的知識綜合起來,提出聚類算法在稅源專業(yè)化管理應(yīng)用的存在問題與對策研究,并研究設(shè)計了聚類算法在稅源專業(yè)化管理應(yīng)用的模型。繼而用SPSS Clementine20.0軟件,同時依照數(shù)據(jù)挖掘的流程建立了標準模型,通過文章選用的指標體系使用K-means聚類方法對WIND分類下的日常消費行業(yè)196家公司進行了分析,論證了聚類算法在稅源專業(yè)化管理應(yīng)用中的適用性。
[Abstract]:Following cloud computing, big data has become one of the hottest concepts in the information technology world. According to the China Internet data Center, the global data volume in 2012 was 2.7ZBP1ZBP106PBP106GB, while the Internet and the mobile Internet recorded individual shopping. Location and other related data are even more massive. So how to use these data has become an increasingly prominent problem. As a tax department collecting money for the country and collecting taxes for the people, There is also the problem of the lack of deep mining and utilization of tax-related data. With the launch of the third issue of the Gold tax, tax data of the tax authorities have changed from a provincial concentration to a large concentration of national data. There are huge amounts of external and internal data in the database of tax departments. However, the analysis and processing of these huge amounts of data by tax authorities are limited to basic applications such as summary, query, contrast, etc. This paper tries to introduce the data mining technology into the work of the tax department so as to make the decision of the tax department more scientific and improve the quality of tax service. At the same time, the focus of the research is on the application of cluster analysis in the specialized management of tax sources. The specialized management of tax sources is the basic work of tax administration, which controls the comprehensive information of tax sources and implements rigorous specialized management methods. Cluster analysis is a kind of data mining technology, which has been widely used in commercial, medical and other fields. Cluster analysis is used in specialized management of tax sources. It can fully improve the effectiveness of tax source classification, at the same time, can enable tax personnel to scientifically formulate tax source management strategies, improve the efficiency of specialized management of tax sources, On the basis of summarizing the research and practice at home and abroad, this paper combs the relevant knowledge of specialized management of tax sources, big data, data mining technology and clustering algorithm, etc. Then, by synthesizing all aspects of knowledge, the paper puts forward the existing problems and countermeasures of the application of clustering algorithm in the specialized management of tax sources, and studies and designs the model of the application of clustering algorithm in the specialized management of tax sources. Then, it uses SPSS Clementine20.0 software to study and design the model of the application of clustering algorithm in the specialized management of tax sources. At the same time, the standard model is established according to the process of data mining, and the K-means clustering method is used to analyze 196 companies in the daily consumer industry under WIND classification through the index system selected in this paper. The applicability of clustering algorithm in the application of tax source specialized management is demonstrated.
【學位授予單位】：集美大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：F812.42

【參考文獻】

相關(guān)期刊論文 前10條

1 歐舸;金曉茜;;淺談稅收大數(shù)據(jù)時代的金稅三期工程[J];中國管理信息化;2017年01期

2 李蘋;劉柯群;;關(guān)于“互聯(lián)網(wǎng)+”背景下創(chuàng)新稅源管理的研究[J];商場現(xiàn)代化;2016年25期

3 于眾;;大數(shù)據(jù)環(huán)境下稅收數(shù)據(jù)深度利用探索[J];經(jīng)濟研究導刊;2016年13期

4 劉尚希;孫靜;;大數(shù)據(jù)治稅的理念、模式及應(yīng)用[J];經(jīng)濟研究參考;2016年09期

5 王曉東;鐘小新;趙建東;陸文君;張海波;;淺議大數(shù)據(jù)管稅[J];中國稅務(wù);2015年12期

6 劉磊;鐘山;;試析大數(shù)據(jù)時代的稅收管理[J];稅務(wù)研究;2015年01期

7 周世佳;;大數(shù)據(jù)思維初探:提出、特征及意義[J];中共山西省直機關(guān)黨校學報;2014年05期

8 譚榮華;焦瑞進;;關(guān)于大數(shù)據(jù)在稅收工作中應(yīng)用的幾點認識[J];稅務(wù)研究;2014年09期

9 崔占如;;聚類分析法在房產(chǎn)稅稅基批量評估中的應(yīng)用[J];河北金融;2014年08期

10 王一民;;稅源專業(yè)化管理模式下數(shù)據(jù)分析系統(tǒng)的應(yīng)用研究[J];稅收經(jīng)濟研究;2014年03期

相關(guān)重要報紙文章 前3條

1 謝永健;;大數(shù)據(jù)：實現(xiàn)稅收現(xiàn)代化的利器[N];中國稅務(wù)報;2014年

2 牟可光;;用好大數(shù)據(jù)，服務(wù)稅收現(xiàn)代化[N];中國稅務(wù)報;2014年

3 程輝;;從挖掘數(shù)據(jù)價值入手增強稅收征管能力[N];中國稅務(wù)報;2013年

相關(guān)碩士學位論文 前6條

1 宋瑜;運用稅收大數(shù)據(jù)加強稅源管理的研究[D];吉林財經(jīng)大學;2016年

2 時待吾;基于數(shù)據(jù)挖掘的企業(yè)欠稅預測研究[D];重慶大學;2016年

3 劉文楠;數(shù)據(jù)挖掘在稅收分析中的應(yīng)用研究[D];財政部財政科學研究所;2014年

4 張佳瑤;基于聚類的數(shù)據(jù)挖掘技術(shù)在稅源監(jiān)控中的應(yīng)用[D];財政部財政科學研究所;2013年

5 翟斌斌;我國商業(yè)銀行的房地產(chǎn)信貸風險研究[D];南京師范大學;2012年

6 錢彥江;大規(guī)模數(shù)據(jù)聚類技術(shù)研究與實現(xiàn)[D];電子科技大學;2009年

，

本文編號：1648103