The computational modeling of deep, robust discourse analysis by integrating abductive reasoning, machine learning, and physical computing
| Project/Area Number |
15H01702
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (A)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Tohoku University |
|---|
Principal Investigator |
Inui Kentaro 東北大学, 情報科学研究科, 教授 (60272689)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
岡崎 直観 東京工業大学, 情報理工学院, 教授 (50601118)
井之上 直也 東北大学, 情報科学研究科, 助教 (80778605)
鈴木 潤 東北大学, 情報科学研究科, 准教授 (80396150)
|
|---|
| Project Period (FY) |
2015-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥42,900,000 (Direct Cost: ¥33,000,000、Indirect Cost: ¥9,900,000)
Fiscal Year 2018: ¥10,790,000 (Direct Cost: ¥8,300,000、Indirect Cost: ¥2,490,000)
Fiscal Year 2017: ¥11,050,000 (Direct Cost: ¥8,500,000、Indirect Cost: ¥2,550,000)
Fiscal Year 2016: ¥11,310,000 (Direct Cost: ¥8,700,000、Indirect Cost: ¥2,610,000)
Fiscal Year 2015: ¥9,750,000 (Direct Cost: ¥7,500,000、Indirect Cost: ¥2,250,000)
|
|---|
| Keywords | 自然言語処理 / 人工知能 / 知能情報学 / 談話解析 / 仮説推論 / 知識獲得 / 物理計算 / 推論 / 知識 |
|---|
| Outline of Final Research Achievements |
This project aimed at the computational modeling of deep, robust discourse analysis that was able to "read between the lines" by integrating abductive reasoning, machine learning, and physical computing. The main achievements are as follows. First, we significantly enhanced the computational capacity of abductive reasoning by formalizing the problem as weighted maximum satisfiability and devising efficient pruning methods. Second, we developed novel methods for large-scale knowledge acquisition from both Web and Wikipedia documents and demonstrated the impact of leveraging acquired knowledge on semantic and discourse analysis. Third, we built and empirically evaluated a computational model that innovatively integrated abductive reasoning and physical simulation to predict risks involved in given traffic scenes. The resources and tools developed in this project are made publicly available on our website.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究は、文章の「行間」を読むことができる計算モデルの構築を目指して論理推論と機械学習の融合を図ったもので、極めて独自性の高い試みであった。仮説論理推論の手法についても既存手法に比べて百倍以上の速度改善を達成しており、実世界データにおける仮説推論の実行可能性を飛躍的に引き上げた。談話解析を物理計算にグラウンディングさせる研究も先例のない挑戦的な試みであったが、その成果として運転シーンからの危険予測技術など実世界へ直接的に組み込める技術の雛形を開発し、「仮説推論による意図理解」というまったく新しい方向性の技術の実例を示した。
|
Report
(5 results)
Research Products
(74 results)
-
-
-
-
-
-
-
-
-
[Journal Article] Cross-lingual Learning-to-Rank with Shared Representations2018
Author(s)
Shota Sasaki, Shuo Sun, Shigehiko Schamoni, Kevin Duh and Kentaro Inui
-
Journal Title
Proceedings of the 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2018)
Volume: -
Related Report
Open Access / Int'l Joint Research
-
[Journal Article] A Melody-conditioned Lyrics Language Model2018
Author(s)
Kento Watanabe, Yuichiroh Matsubayashi, Satoru Fukayama, Masataka Goto, Kentaro Inui and Tomoyasu Nakano
-
Journal Title
Proceedings of the 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2018)
Volume: -
Related Report
Peer Reviewed / Open Access
-
[Journal Article] Modeling Storylines in Lyrics2017
Author(s)
Kento Watanabe, Yuichiroh Matsubayashi, Kentaro Inui, Satoru Fukayama, Tomoyasu Nakano, and Masataka Goto
-
Journal Title
IEICE Transactions on Information and Systems
Volume: Vol.E101-D, No.4
NAID
Related Report
Peer Reviewed / Open Access
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
[Journal Article] Recognizing Open-vocabulary Relations between Objects in Images2016
Author(s)
Masayasu Muraoka, Sumit Maharjan, Masaki Saito, Kota Yamaguchi, Naoaki Okazaki, Takayuki Okatani and Kentaro Inui
-
Journal Title
In Proceedings of the 30th Pacific Asia Conference on Language, Information and Computation (PACLIC 30)
Volume: なし
Pages: 339-248
NAID
Related Report
Peer Reviewed / Open Access / Acknowledgement Compliant
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
[Presentation] Suspicious News Detection Using Micro Blog Text2018
Author(s)
Tsubasa Tagami, Hiroki Ouchi, Hiroki Asano, Kazuaki Hanawa, Kaori Uchiyama, Kaito Suzuki, Kentaro Inui, Atsushi Komiya, Atsuo Fujimura, Ryo Yamashita, Hitofumi Yanai and Akinori Machino
Organizer
Proceedings of the 32nd Pacific Asia Conference on Language, Information and Computation (PACLIC 32)
Related Report
Int'l Joint Research
-
-
-
-
-
-
-
[Presentation] A Melody-conditioned Lyrics Language Model2018
Author(s)
Kento Watanabe, Yuichiroh Matsubayashi, Satoru Fukayama, Masataka Goto, Kentaro Inui and Tomoyasu Nakano
Organizer
Proceedings of the 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2018)
Related Report
Int'l Joint Research
-
-
-
-
-
-
-
-
-
-
-
-
-
