| Project/Area Number |
16K13189
|
|---|
| Research Category |
Grant-in-Aid for Challenging Exploratory Research
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Art at large
|
|---|
| Research Institution | Rikkyo University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
尾鼻 崇 中部大学, 人文学部, 講師 (00516833)
曹 慶鎬 立教大学, 社会学部, 教育研究コーディネーター (20762892)
|
|---|
| Project Period (FY) |
2016-04-01 – 2018-03-31
|
| Project Status |
Completed (Fiscal Year 2017)
|
| Budget Amount *help |
¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2017: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2016: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
|
|---|
| Keywords | テキストマイニング / 機械学習 / Python / Word2vec / Twitter / ヘイト・スピーチ / 選挙 / クラスター分析 / ヘイトスピーチ / フェイクニュース / ソフトウェア・スタディーズ / 表象文化 |
|---|
| Outline of Final Research Achievements |
In this research, we collected a large amount of Twitter data, analyzed data using statistical analysis tools and machine learning in Python, and clarified what kinds of groups were present and what kind of words were written.
Twitter is a popular SNS in Japan, but in recent years it has been said that there are many negative contents such as slander and it is regarded as a problem. In addition to analyzing negative contents, we also analyzed positive contents. We collected Twitter data on the Kumamoto earthquake that occurred in April 2016 and data on Tokyo gubernatorial election's election held in July 2016 on a large scale and analyzed them.
As a result, we could clarify that because of the design of User Interface of Twitter, negative contents only felt a lot, there is only a few people wrote negative words.
|
