2021 Fiscal Year Final Research Report
Research on Interview Dialogue to obtain User's good points by Conversation Robot
| Project/Area Number |
19K12174
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 61050:Intelligent robotics-related
|
|---|
| Research Institution | Kagawa National College of Technology |
|---|
Principal Investigator |
Sasayama Manabu 香川高等専門学校, 情報工学科, 准教授 (60508232)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
松本 和幸 徳島大学, 大学院社会産業理工学研究部(理工学域), 准教授 (90509754)
|
|---|
| Project Period (FY) |
2019-04-01 – 2022-03-31
|
| Keywords | インタビュー対話 / 発話意図タグ / インタビュー対話コーパス |
|---|
| Outline of Final Research Achievements |
In this study, we constructed an interview dialogue corpus tagged with utterance intention. The interview dialogue corpus was collected from a TV dialogue program. 689 interview dialogues were collected. We designed utterance intention tags. We defined 18 types of tags: 4 types of first-level tags and 14 types of second-level tags. We assigned one first-level tag and one second-level tag to each utterance in the 30 dialogues (14761 utterances) in which one person was a guest. Ten male workers aged between 18 and 24 years old were assigned by three or five persons per dialogue.
|
| Free Research Field |
自然言語処理
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究においてインタビュー対話における質問とその回答，相槌，感想を区別したコーパスを構築できた．本コーパスを利用することで，話題毎の質問と回答を対話システムの応答に利用でき，相槌の挿入のタイミングを研究することにもつながる．特に，ユーザの回答として入力した文と構造的に似ている回答発話を検索することができるため，次に相槌を入れるか判定できたり，話題毎の深さなどを判定できたりし，その知見を発表することで対話研究の進展に貢献できる．
|
