2023 Fiscal Year Final Research Report
A study on integrated basic technologies for inferring emotions of tweet posters
| Project/Area Number |
20K12085
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 62020:Web informatics and service informatics-related
|
|---|
| Research Institution | Chiba Institute of Technology |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
灘本 明代 甲南大学, 知能情報学部, 教授 (30359103)
|
|---|
| Project Period (FY) |
2020-04-01 – 2024-03-31
|
| Keywords | ツイート感情 / 顔文字感情 / 感情推測 |
|---|
| Outline of Final Research Achievements |
In this study, we focused on the ten basic emotion types of "sad, dislike, relief, fear, elation, like, joy, surprise, anger, and ashame," and quantatively represented the emotions of tweets with no emoticons, emotions of emoticons, and emotions of tweets with emoticons in the form of ten-dimensional vectors. We formulated the influence that the emotions felt from emoticons have on the emotions felt from tweets, and then we integrated the results with our proposed method that quantatively infers emotions of tweets. In addition, we expanded our integrated method to handle more fine-grained negative emotion types (dissatisfaction, anxiety, disappointment, anguish, hesitation, pain, prejudice, and discrimination), targeting more than 10,000 negative tweets related to COVID-19.
|
| Free Research Field |
データ工学
|
| Academic Significance and Societal Importance of the Research Achievements |
ツイート投稿者（現，X投稿者）の感情推測において，ツイートの感情，顔文字の感情，顔文字付きツイートの感情という3つの要素すべてを，10個の基本感情を構成要素とする10次元のベクトルとして扱うという研究は，これまでなかった．ツイートの感情だけでなく，顔文字の感情もベクトル化することで，顔文字の感情がツイートの感情に与える影響を明らかにするだけでなく，顔文字付きツイートの感情をより高精度に推測できるようになった．既存の研究（トレンド分析やニュース記事推薦，ユーザの印象選好の可視化，SNS等でのフォロイー推薦など）に応用することで，よりリッチなWebサービスの提供も可能になるものと考えられる．
|
