Detection of cheating in computer-based testing using biometrics
| Project/Area Number |
25350355
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Educational technology
|
|---|
| Research Institution | Seikei University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SUZUKI Satoshi 成蹊大学, 理工学部, 助教 (70516377)
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
MURAMATSU Daigo 大阪大学, 産業科学研究所, 特任講師 (00386624)
|
|---|
| Project Period (FY) |
2013-04-01 – 2016-03-31
|
| Project Status |
Completed (Fiscal Year 2015)
|
| Budget Amount *help |
¥4,810,000 (Direct Cost: ¥3,700,000、Indirect Cost: ¥1,110,000)
Fiscal Year 2015: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2014: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
Fiscal Year 2013: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
|---|
| Keywords | オンライン試験 / バイオメトリクス / 不正行為検出 / 替え玉受験 / 聴覚利用型不正行為 / 自宅受験 / 本人認証 / 不正行為防止 / サウンドマスキング / タブレット / マルチタッチジェスチャ / 行動特徴 / タブレット型端末 |
|---|
| Outline of Final Research Achievements |
Methods of detecting several kinds of cheating using biometrics are proposed to augment reliability of unproctored online exams such as computerized take-home testing.In the first subtheme, a method of detecting impersonations for exams that use tablet computers is examined. An authentication technique that utilizes touch gesture features of the examinee is developed. The gestures are obtained using the camera embedded in the tablet device by taking from the side or from the top of the hand. More than 90% were correctly authenticated in the experiment. In the second subtheme, detection of cheating using acoustic device is addressed. A detection method using gaze motion is proposed, and its effectiveness is verified in the experiment. The result indicates that the gaze behaviors between a reading task, and a dual task of pretending reading a dummy text while carrying out a listening task are different. 77% were correctly authenticated without information dependent on examinees.
|
Report
(4 results)
Research Products
(14 results)
