Detecting phrase similarity based on compositional symbolic computation and statistical knowledge acquisition using a large-scale corpus.

• Project/Area Number: 15K16045
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Intelligent informatics
• Research Institution: Institute of Physical and Chemical Research (2018); Tohoku University (2015-2017)
• Principal Investigator: Matsubayashi Yuichiroh 国立研究開発法人理化学研究所, 革新知能統合研究センター, 研究員 (20582901)
• Project Period (FY): 2015-04-01 – 2019-03-31
• Project Status: Completed (Fiscal Year 2018)
• Budget Amount: ¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000); Fiscal Year 2017: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000); Fiscal Year 2016: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000); Fiscal Year 2015: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
• Keywords: 自然言語処理 / 人工知能 / 項構造解析 / 談話 / 知識 / 文の意味構造解析 / 省略解析 / 意味構造解析 / 文章生成 / 項構造アノテーション / 多言語翻訳 / 方言翻訳 / 概念構造 / 類犠牲認識 / 分散意味論 / 選択選考モデル / 時間関係認識

Outline of Final Research Achievements

In this research, in order to establish a semantic parser that can handle various linguistic expressions robustly, we have aimed to build models that can handle semantics of rather longer expressions than words, such as phrases, sentences, and discourse structures of multiple sentences. Specifically, employing distributed word representations as a key technology, we (1) improved semantic structure analysis models for phrases including multiple events, (2) improved semantic models that can handle inter-sentential semantic structures, and (3) proposed computational models that incorporate linguistic and non-linguistic multimodal structures. The contribution in this research project was presented in 26 publications including 11 peer-reviewed papers.

Academic Significance and Societal Importance of the Research Achievements

(1)で開発した意味解析器では、現状の日本語解析における重大なボトルネックである「省略された内容を補う」解析の精度を当初の40%強から60%弱まで飛躍的に向上させ実用レベルに近づけた。開発したシステムは一般公開し、実世界テキスト解析に適用可能である。(2)では文の解析に先行文脈の情報を利用するという近年取り組みが減っていたアイデアに再注目、ハイライトした。このアイデアに基づく研究の数は徐々に増えている。(3) では、音楽、絵本の２つの題材を取り上げ、言語の構造とそれを取り巻く情報の構造を相互に考慮することが言語構造の推定に重要な役割を果たすことを示し、分野での先行事例としての役割を果たした。

Research Products

• [Journal Article] Modeling Storylines in Lyrics (2018)
  • Author(s): Kento WATANABE, Yuichiroh MATSUBAYASHI, Kentaro INUI, Satoru FUKAYAMA, Tomoyasu NAKANO, Masataka GOTO
  • Journal Title: IEICE Transactions on Information and Systems Volume: E101.D Issue: 4 Pages: 1167-1179
  • DOI: 10.1587/transinf.2017EDP7188
  • NAID: 130006602302
  • ISSN: 0916-8532, 1745-1361
  • Peer Reviewed / Open Access
• [Journal Article] 3.3 述語項構造解析 (2016)
  • Author(s): 松林優一郎
  • Journal Title: 情報処理 Volume: Vol.57, No.1 Pages: 14-15
• [Journal Article] 日本語文書内で表現される事象間の時間的な順序関係の推定 (2015)
  • Author(s): 稲田和明, 松林優一郎, 乾健太郎
  • Journal Title: 情報処理学会論文誌 Volume: 56(10) Pages: 2054-2071
  • NAID: 170000130757
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Issues on Annotation Guidelines for Japanese Predicate-Argument Structures (2014)
  • Author(s): 松林優一郎, 中山周, 乾健太郎
  • Journal Title: Journal of Natural Language Processing Volume: 21 Issue: 2 Pages: 333-377
  • DOI: 10.5715/jnlp.21.333
  • NAID: 130004566475
  • ISSN: 1340-7619, 2185-8314
  • Peer Reviewed / Open Access / Acknowledgement Compliant
• [Presentation] Verb arguments in Japanese picture books (2019)
  • Author(s): Naho Orita, Asumi Suzuki, Yuichiro Matsubayashi.
  • Organizer: The 41st Annual Meeting of the Cognitive Science Society
  • Int'l Joint Research
• [Presentation] 前方文脈の埋め込みを利用した日本語述語項構造解析 (2019)
  • Author(s): 今野颯人, 松林優一郎, 大内啓樹, 清野舜, 乾健太郎
  • Organizer: 言語処理学会第25回年次大会
• [Presentation] Multi-dialect Neural Machine Translation and Dialectometry (2018)
  • Author(s): Kaori Abe, Yuichiroh Matsubayashi, Naoaki Okazaki, and Kentaro Inui
  • Organizer: The 32nd Pacific Asia Conference on Language, Information and Computation
  • Int'l Joint Research
• [Presentation] Distance-Free Modeling of Multi-Predicate Interactions in End-to-End Japanese Predicate-Argument Structure Analysis (2018)
  • Author(s): Yuichiroh Matsubayashi and Kentaro Inui
  • Organizer: The 27th International Conference on Computational Linguistics
  • Int'l Joint Research
• [Presentation] A Melody-conditioned Lyrics Language Model (2018)
  • Author(s): Kento Watanabe, Yuichiroh Matsubayashi, Satoru Fukayama, Masataka Goto, Kentaro Inui and Tomoyasu Nakano
  • Organizer: The 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
• [Presentation] A Melody-conditioned Lyrics Language Model (2018)
  • Author(s): Kento Watanabe, Yuichiroh Matsubayashi, Satoru Fukayama, Masataka Goto, Kentaro Inui and Tomoyasu Nakano
  • Organizer: The 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)
  • Int'l Joint Research
• [Presentation] 複数の述語間の関係を考慮したEnd-to-End 日本語述語項構造解析 (2018)
  • Author(s): 松林優一郎, 乾健太郎
  • Organizer: 言語処理学会第24回年次大会
• [Presentation] ニューラルネットを用いた多方言の翻訳と類型分析 (2018)
  • Author(s): 阿部香央莉, 松林優一郎, 岡崎直観, 乾健太郎
  • Organizer: 言語処理学会第24回年次大会
• [Presentation] 絵本述語項構造コーパスの設計と構築 (2018)
  • Author(s): 折田奈甫, 石井啓太, 鈴木あすみ, 松林優一郎
  • Organizer: 言語処理学会第24回年次大会
• [Presentation] メロディ条件付き歌詞言語モデル (2018)
  • Author(s): 渡邉研斗, 松林優一郎, 深山覚, 乾健太郎, 後藤真孝, 中野倫靖
  • Organizer: 言語処理学会第24回年次大会
• [Presentation] ニューラルネットワークによる日本語述語項構造解析の素性の汎化 (2017)
  • Author(s): 松林優一郎, 乾健太郎
  • Organizer: 言語処理学会
  • Place of Presentation: 筑波大学（つくば市）
  • Year and Date: 2017-03-14
• [Presentation] LyriSys: An Interactive Support System for Writing Lyrics Based on Topic Transition (2017)
  • Author(s): Kento Watanabe, Yuichiroh Matsubayashi, Kentaro Inui, Tomoyasu Nakano, Satoru Fukayama and Masataka Goto
  • Organizer: Annual Meeting of the Intelligent User Interfaces Community
  • Place of Presentation: Limassol, Cyprus
  • Year and Date: 2017-03-13
  • Int'l Joint Research
• [Presentation] Revisiting Design Issues of Local Models for Japanese Predicate-Argument Structure Analysis (2017)
  • Author(s): Yuichiroh Matsubayashi and Kentaro Inui
  • Organizer: The 8th International Joint Conference on Natural Language Processing (IJCNLP)
  • Int'l Joint Research
• [Presentation] 日本語メタファー表現に対する分散表現の評価と考察 (2017)
  • Author(s): 菊地凜, 松林優一郎, 乾健太郎
  • Organizer: NLP若手の会 シンポジウム
• [Presentation] Dependency-based Compositional Semantics Vectorに基づく文脈付き選択選好モデル (2017)
  • Author(s): 高橋諒, 松林優一郎, 中山周, 田然, 乾健太郎
  • Organizer: NLP若手の会 シンポジウム
• [Presentation] メロディと歌詞の相関に基づく自動歌詞生成 (2017)
  • Author(s): 渡邉研斗、松林優一郎、深山覚、中野倫靖、後藤真孝、乾健太郎
  • Organizer: 情報処理学会第231回 自然言語処理研究会
• [Presentation] Modeling Context-sensitive Selectional Preference with Distributed Representations (2016)
  • Author(s): Naoya Inoue, Yuichiro Matsubayashi, Masayuki Ono, Naoaki Okazaki and Kentaro Inui
  • Organizer: International Conference on Computational Linguistics
  • Place of Presentation: Osaka International Convention Center, Japan
  • Year and Date: 2016-12-13
  • Int'l Joint Research
• [Presentation] Modeling Discourse Segments in Lyrics Using Repeated Patterns (2016)
  • Author(s): Kento Watanabe, Yuichiroh Matsubayashi, Naho Orita, Naoaki Okazaki, Kentaro Inui, Satoru Fukayama, Tomoyasu Nakano, Jordan B. L. Smith and Masataka Goto
  • Organizer: International Conference on Computational Linguistics
  • Place of Presentation: Osaka International Convention Center, Japan
  • Year and Date: 2016-12-13
  • Int'l Joint Research
• [Presentation] 分散表現による文脈情報を用いた選択選好モデル (2016)
  • Author(s): 大野雅之, 井之上直也, 松林優一郎, 岡崎直観, 乾健太郎
  • Organizer: 言語処理学会第22回年次大会
  • Place of Presentation: 東北大学
  • Year and Date: 2016-03-08
• [Presentation] 分散表現に基づく選択選好モデルの文脈化 (2016)
  • Author(s): 大野雅之, 井之上直也, 松林優一郎, 岡崎直観, 乾健太郎
  • Organizer: 情報処理学会 自然言語処理研究会報告
  • Place of Presentation: 株式会社ミクシィ 住友不動産渋谷ファーストタワー 7F
  • Year and Date: 2016-01-22
• [Book] 小川芳樹, 長野明子, 菊地朗 (編), コーパスからわかる言語変化・変異と言語理論, Part V (2016)
  • Author(s): 福原裕一, 松林優一郎, 乾健太郎
  • Publisher: 開拓社