| Project/Area Number |
25280118
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Partial Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Library and information science/Humanistic social informatics
|
|---|
| Research Institution | Kyushu University |
|---|
Principal Investigator |
Ishita Emi 九州大学, 附属図書館, 准教授 (50364815)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
冨浦 洋一 九州大学, システム情報科学研究院, 教授 (10217523)
高山 泰博 徳山工業高等専門学校, その他部局等, 教授 (30565841)
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
OGA Toru 九州大学, 法学学術院, 准教授 (90445718)
|
|---|
| Research Collaborator |
Fleischmann Kenneth R. テキサス大学, オースティン校
Oard Douglas W. メリーランド大学, カレッジパーク校
Cheng An-Shou 国立中山大学
|
|---|
| Project Period (FY) |
2013-04-01 – 2018-03-31
|
| Project Status |
Completed (Fiscal Year 2017)
|
| Budget Amount *help |
¥11,180,000 (Direct Cost: ¥8,600,000、Indirect Cost: ¥2,580,000)
Fiscal Year 2016: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
Fiscal Year 2015: ¥3,250,000 (Direct Cost: ¥2,500,000、Indirect Cost: ¥750,000)
Fiscal Year 2014: ¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
Fiscal Year 2013: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
|---|
| Keywords | 価値観推定 / 内容分析 / 価値観 / 自動分類 / テストコレクション / 国際研究者交流 / 価値観分析 / 国際情報交換 |
|---|
| Outline of Final Research Achievements |
This report describes the development and evaluation of a "Latent Variable Model" (LVM) classifier that detects evidence for specific human values at either the passage or full-document level. To determine suitability for a specific application, an extrinsic evaluation method reflecting the use of the classification results for content analysis was performed. To extend the utility of the technique, the relationship between words and human value was analyzed to discern patterns of language use. To characterize the versatility of the classifier, new test collections containing newspaper articles about the Fukushima Daiichi nuclear power disaster were developed in which human values were manually assigned as a gold standard at document level. Results obtained on these collections has informed the design of future test collections in which human annotation will be performed at sentence level and multiple categories.
|
