應用逐步迴歸、決策樹、約略集合及類神經網路於偵測企業舞弊

CCUR > College of Business > Department of Accounting & Graduate Institute of Accounting > Thesis > Item 987654321/28246

Please use this identifier to cite or link to this item: https://irlib.pccu.edu.tw/handle/987654321/28246

Title:	應用逐步迴歸、決策樹、約略集合及類神經網路於偵測企業舞弊 Applying Stepwise Regression, Decision Tree, Rough Set Theory, and Artificial Neural Network for Detecting Fraud of Enterprises
Authors:	游謹安
Contributors:	會計學系
Keywords:	舞弊資料探勘決策樹約略集合類神經網路
Date:	2014
Issue Date:	2014-09-26 15:49:47 (UTC+8)
Abstract:	近年來隨著舞弊案件不斷增加，當公司發生重大弊案時，不僅使公司本身受到傷害，更造成投資人的重大損失，使得社會必須極大的成本來彌補其所造成的傷害。過去有關企業舞弊的文獻中，主要使用傳統的迴歸模式為主，而近年來有許多學者使用資料探勘來偵測企業舞弊，都獲得相當不錯的準確率，但整體文獻還不夠完整。故本研究第一階段以傳統的逐步迴歸法和資料探勘中的卡方自動交叉驗證(CHAID)及約略集合篩選出重要變數，配合決策樹C5.0及類神經網路分別建構分類模型並進行比較，變數方面則採用財務及非財務變數，希望能建立一套更為有效的企業舞弊偵測之工具。本研究之研究對象為2003年至2013年間，41家發生企業舞弊及123家非企業舞弊之公司。研究結果發現，約略集合搭配類神經網路之企業舞弊偵測模型能作為協助審計人員於查核過程中偵測企業舞弊之工具，及提供審計人員及投資大眾作為重要決策之參考。 More and more frauds have occurred in recent years. When corporations involve in critical frauds, the corporation itself and investors suffer from all such frauds. As a result, the society incurs huge costs in order to make up for the loss caused by the frauds. Most of the fraud-related literatures studied frauds using the conventional re-gression model. In recent years, however, many researchers detected frauds using data mining method with satisfactory accuracy. However, not enough literatures are available at this moment. Therefore, this study attempted to identify critical variables using the conven-tional stepwise regression method, the Chi-square automatic Interaction Detection (CHAID) designed for data mining, as well as rough set. This study adopted financial variables and non-financial variables in order to construct an effective tool to detect frauds. Moreover, this study focused on 41 enterprises involved in corporate frauds and 123 enterprises not involved in corporate frauds in 2003-2013. According to the research results, the fraud detection model constructed with ar-tificial neural network together with rough set is sufficient to detect fraud for auditors, and is a perfect tool for the auditors and investors whenever they have to reach major decisions.
Appears in Collections:	[Department of Accounting & Graduate Institute of Accounting] Thesis

Files in This Item:

There are no files associated with this item.

View Licence

Loading...