學術產出-Journal Articles

Article View/Open

Publication Export

Google ScholarTM

政大圖書館

Citation Infomation

題名 資料探勘技術在繼續經營疑慮意見診斷模型之應用
Going Concern Opinion: Application of Data Mining Technologies
作者 盧鈺欣
林昱成
林育伶
Lu, Yu-Hsin
Lin, Yu-Cheng
Liao, Jung-Ling
關鍵詞 繼續經營疑慮意見 ; 資料探勘技術 ; 特徵選擇 ; 分類技術
Going concern opinion ; Data mining technologies ; Feature selection ; Classifier technique
日期 2016-07
上傳時間 15-Nov-2017 16:01:10 (UTC+8)
摘要 會計師決定是否出具繼續經營疑慮意見時,涉及專業判斷且考量因素眾多與複雜。因此,評估公司繼續經營假設是否有重大疑慮的分析性資訊對會計師而言非常重要。本文之目的係以資料探勘技術建構繼續經營疑慮意見診斷模型,並提供會計師有用之決策資訊,藉以輔助其評估對受查客戶出具繼續經營疑慮意見書之依據。首先,本文利用特徵選擇工具自眾多影響會計師出具繼續經營疑慮意見的相關變數中,篩選出6 個重要影響因素。再輔以分類技術-決策樹建構繼續經營疑慮意見診斷模型,並產出決策表供會計師參酌。實證結果顯示,本文決策表所提供之10 條分類規則,能有效區別繼續經營疑慮意見書類型,其預測準確率高達91.35%,有助於會計師評估繼續經營疑慮意見時之參考依據,降低審計風險。
The auditors` going concern opinion usually involves complex professional judgment and considerations. Therefore, information that may raise auditors` substantial doubts as to whether a going-concern opinion should be issued is important during the audit process. This study adopts the data mining technology to build up a going concern diagnostic model from which the auditors can obtain useful information to assess clients’ ability of remaining as a going concern. Specifically, the auditors’ going concern opinion is determined by considering six critical factors extracted from a feature selection tool and a decision table created by a diagnostic model built from a decision tree. The empirical results indicate that the 10 classification rules generated by the decision table can effectively distinguish different types of going concern audit reports with a prediction accuracy of 91.35%. Overall, this decision table facilitates the auditors in assessing clients` likelihood of continuing as a going concerns and, therefore, reducing audit risk.
關聯 會計評論, 63, 77-108
資料類型 article
DOI http://dx.doi.org/10.6552/JOAR.2016.63.3
dc.creator (作者) 盧鈺欣zh_TW
dc.creator (作者) 林昱成zh_TW
dc.creator (作者) 林育伶zh_TW
dc.creator (作者) Lu, Yu-Hsinen_US
dc.creator (作者) Lin, Yu-Chengen_US
dc.creator (作者) Liao, Jung-Lingen_US
dc.date (日期) 2016-07-
dc.date.accessioned 15-Nov-2017 16:01:10 (UTC+8)-
dc.date.available 15-Nov-2017 16:01:10 (UTC+8)-
dc.date.issued (上傳時間) 15-Nov-2017 16:01:10 (UTC+8)-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/114780-
dc.description.abstract (摘要) 會計師決定是否出具繼續經營疑慮意見時,涉及專業判斷且考量因素眾多與複雜。因此,評估公司繼續經營假設是否有重大疑慮的分析性資訊對會計師而言非常重要。本文之目的係以資料探勘技術建構繼續經營疑慮意見診斷模型,並提供會計師有用之決策資訊,藉以輔助其評估對受查客戶出具繼續經營疑慮意見書之依據。首先,本文利用特徵選擇工具自眾多影響會計師出具繼續經營疑慮意見的相關變數中,篩選出6 個重要影響因素。再輔以分類技術-決策樹建構繼續經營疑慮意見診斷模型,並產出決策表供會計師參酌。實證結果顯示,本文決策表所提供之10 條分類規則,能有效區別繼續經營疑慮意見書類型,其預測準確率高達91.35%,有助於會計師評估繼續經營疑慮意見時之參考依據,降低審計風險。zh_TW
dc.description.abstract (摘要) The auditors` going concern opinion usually involves complex professional judgment and considerations. Therefore, information that may raise auditors` substantial doubts as to whether a going-concern opinion should be issued is important during the audit process. This study adopts the data mining technology to build up a going concern diagnostic model from which the auditors can obtain useful information to assess clients’ ability of remaining as a going concern. Specifically, the auditors’ going concern opinion is determined by considering six critical factors extracted from a feature selection tool and a decision table created by a diagnostic model built from a decision tree. The empirical results indicate that the 10 classification rules generated by the decision table can effectively distinguish different types of going concern audit reports with a prediction accuracy of 91.35%. Overall, this decision table facilitates the auditors in assessing clients` likelihood of continuing as a going concerns and, therefore, reducing audit risk.en_US
dc.format.extent 664171 bytes-
dc.format.mimetype application/pdf-
dc.relation (關聯) 會計評論, 63, 77-108zh_TW
dc.subject (關鍵詞) 繼續經營疑慮意見 ; 資料探勘技術 ; 特徵選擇 ; 分類技術zh_TW
dc.subject (關鍵詞) Going concern opinion ; Data mining technologies ; Feature selection ; Classifier techniqueen_US
dc.title (題名) 資料探勘技術在繼續經營疑慮意見診斷模型之應用zh_TW
dc.title (題名) Going Concern Opinion: Application of Data Mining Technologiesen_US
dc.type (資料類型) article-
dc.identifier.doi (DOI) 10.6552/JOAR.2016.63.3-
dc.doi.uri (DOI) http://dx.doi.org/10.6552/JOAR.2016.63.3-