| Project/Area Number |
15K16044
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Kitami Institute of Technology |
|---|
Principal Investigator |
|
|---|
| Research Collaborator |
MASUI Fumito 北見工業大学, 工学部, 准教授 (80324549)
|
|---|
| Project Period (FY) |
2015-04-01 – 2018-03-31
|
| Project Status |
Completed (Fiscal Year 2017)
|
| Budget Amount *help |
¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000)
Fiscal Year 2017: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2016: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2015: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
|
|---|
| Keywords | 自然言語処理 / ネットいじめ検出 / ネットいじめ対策 / パターン抽出 / ネットいじめ |
|---|
| Outline of Final Research Achievements |
In this research, I developed a technology for automatic detection of cyberbullying with particular focus on both victims and perpetrators. As a research outcome, I first constructed three corpora: (A) general blog corpus, (B) harmful expression corpus, and (C) depressive expression corpus. Further, I investigated statistical characteristics of the collected data. Next, the harmful expression detection module and depression expression detection module were designed, implemented and experimental evaluations were carried out. Specifically, I compared 13 types of Natural Language Processing methods (morphological analysis, syntactic parsing, named entity extraction, etc.) and multiple Artificial Intelligence methods (13 classifiers) were applied and compared. Ultimately, the method with the highest classification performance of 93.5% was experimentally confirmed.
|
