| Project/Area Number |
15K21514
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Learning support system
Educational technology
|
|---|
| Research Institution | Osaka Sangyo University |
|---|
Principal Investigator |
Ohno Asako 大阪産業大学, 工学部, 講師 (90550369)
|
|---|
| Project Period (FY) |
2015-04-01 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥2,600,000 (Direct Cost: ¥2,000,000、Indirect Cost: ¥600,000)
Fiscal Year 2017: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2016: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2015: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | 記述スタイル特徴 / 作成者認証 / レポート盗用発見 / 記述スタイルモデル / 教授学習支援システム / 知的学習支援システム / 特徴抽出 / 類似性検出 / 教育工学 / 知的学習システム / 隠れマルコフモデル / 授業支援システム / 記述における癖の抽出 / 異常検知 / 系列パターンマイニング / 機械学習 / モデル化 |
|---|
| Outline of Final Research Achievements |
I proposed a new plagiarism detection method for Japanese document based on the unique viewpoint of representing an author's writing style and performing author identification and implemented the method in the plagiarism detection system for Japanese academic reports.
The method measures similarity between the same author's previous writing style and the present one. For example, if the writing style extracted from student A's report document is different from his/her previous ones, we assume that there is a possibility of plagiarism.
I confirmed high performance for plagiarism detection from the result of evaluation experiment, though it need more improvement to be used in practice. Parameters of the trained writing models provided a variety of information that can be used to differentiate authors' writing style. The method also has a potential of achieve less burdensome plagiarism detection according to the results of questionnaire investigations.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究では「授業課題レポート文書」の性質や，「授業における盗用発見」という目的に特化し，人の行う「レポート文書の作成者の推定」に倣った，独自の類似性検出手法を提案している．「同一作成者のこれまでの書き方」をもとに，対象文書の作成者認証を行うというこれまでにないアプローチは，誤判定リスクの軽減にも貢献し，盗用発見に関わる教員・学生の精神的な負担軽減にもつながることが期待される．
|
