| Project/Area Number |
18K01923
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 07100:Accounting-related
|
|---|
| Research Institution | Hosei University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2018-04-01 – 2023-03-31
|
| Project Status |
Completed (Fiscal Year 2022)
|
| Budget Amount *help |
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2022: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2021: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2020: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2019: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2018: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
|
|---|
| Keywords | 不正会計 / 機械学習 / テキストマイニング / XBRL / 監査における機械学習の応用 / 有価証券報告書のコーパス / 形態素解析 / 会計不正 / ディープラーニング / 不正発見 / 深層学習 / 畳み込みニューラルネットワーク |
|---|
| Outline of Final Research Achievements |
The purpose of this study is to examine whether machine learning or text mining methods can increase the likelihood of detection of accounting fraud over the conventional analysis method. The entire securities report data of all listed companies uploaded to the EDINET System is available in XBRL format from the fiscal years of 2013. As a result,we have access not only to financial information, but also to narrative information. The first step of this study was obtaining all of these data, pre-processing the data, and accumulating the data for analysis.
Then, using the database of the "Companies Disclosing Improper Accounting Survey (2008-2021)" by Tokyo Shoko Research, a comparison of data with a group of companies of similar size in the same industry was conducted to grasp the characteristics of the financial data of companies that have committed fraud. However, while no major differences of financial data and word and phrases characteristics of each company.
|
| Academic Significance and Societal Importance of the Research Achievements |
これまで不正会計に関する研究は、財務分析を中心とした研究はそれなりの蓄積があったが、ナラティブ情報のような非財務情報に対する分析については十分な蓄積があるとはいえない。そこに近年発展の著しい機械学習やテキストマイニングの手法を応用して不正会計を行った企業のもつ特徴をとらえようとする追加的貢献をもたらすことに、本研究の学術的意義がある。本研究の結果はあまり芳しいものではなかったが、今後より進展すれば、不正会計を行っている企業を捉えることが可能となり、不正会計によってもたらされる損失を回避できる可能性が高まる点において、社会的意義があるといえる。
|
