• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Development of Incremental Knowledge Acquisition in Dialogue Systems Based on Knowledge Model Generalization

Research Project

Project/Area Number 19H04171
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Review Section Basic Section 61030:Intelligent informatics-related
Research InstitutionOsaka University

Principal Investigator

KOMATANI Kazunori  大阪大学, 産業科学研究所, 教授 (40362579)

Co-Investigator(Kenkyū-buntansha) 武田 龍  大阪大学, 産業科学研究所, 准教授 (20749527)
林 克彦  群馬大学, 情報学部, 准教授 (50725794)
Project Period (FY) 2019-04-01 – 2022-03-31
Project Status Completed (Fiscal Year 2022)
Budget Amount *help
¥17,290,000 (Direct Cost: ¥13,300,000、Indirect Cost: ¥3,990,000)
Fiscal Year 2021: ¥5,330,000 (Direct Cost: ¥4,100,000、Indirect Cost: ¥1,230,000)
Fiscal Year 2020: ¥5,460,000 (Direct Cost: ¥4,200,000、Indirect Cost: ¥1,260,000)
Fiscal Year 2019: ¥6,500,000 (Direct Cost: ¥5,000,000、Indirect Cost: ¥1,500,000)
Keywords対話システム / 知識グラフ補完 / ドメイン知識獲得 / 未知概念 / 暗黙的確認 / 質問選択 / ユーザ心象 / 未知語
Outline of Research at the Start

賢い対話システムを作るには知識が不可欠ですが、完璧な知識を作り込むのはほぼ不可能です。このため、話の最中に、知識の欠けを補う技術が必要です。本研究では、システムの知識表現として知識グラフを用い、その上で知識グラフ補完と呼ばれる未知の関係を推測する技術などを用いて、適切な質問内容を選べるようにします。これにより「話すほどに賢くなる対話システム」の実現に向けた基盤技術の確立を目指します。

Outline of Final Research Achievements

Questions to acquire knowledge about unknown words that appear in a dialogue need to be asked naturally without discouraging the user from continuing the dialogue. In this research project, we experimentally demonstrated the relationship between the knowledge graph completion score and the accuracy of question content when selecting questions using the knowledge graph completion results. Furthermore, we demonstrated the effectiveness of the framework of the question selection based on the score by conducting human evaluation by crowdsourcing.

Academic Significance and Societal Importance of the Research Achievements

対話システムが対話中に知識を獲得するのは,飽きられず長期にわたる関係性を保てるシステムを構築するうえで重要なトピックである.従来は保持する知識から応答を生成する方法の研究が多いが,本研究では,知識グラフが不完全であることに着目し,質問に対するユーザの印象も考慮しながら,その欠損値をシステムが獲得する.未知語問題や未知の知識など「システムができないこと」に正面から焦点を当てる本研究の取組みは実用上重要である.対話中の知識獲得の研究は未だ世界的にも少ない.

Report

(4 results)
  • 2022 Final Research Report ( PDF )
  • 2021 Annual Research Report
  • 2020 Annual Research Report
  • 2019 Annual Research Report
  • Research Products

    (35 results)

All 2022 2021 2020 2019 Other

All Journal Article (17 results) (of which Peer Reviewed: 17 results,  Open Access: 10 results) Presentation (16 results) (of which Int'l Joint Research: 1 results,  Invited: 1 results) Remarks (1 results) Patent(Industrial Property Rights) (1 results)

  • [Journal Article] Lexical Acquisition during Dialogues through Implicit Confirmation2022

    • Author(s)
      Komatani Kazunori、Ono Kohei、Takeda Ryu、Nichols Eric、Nakano Mikio
    • Journal Title

      Dialogue & Discourse

      Volume: 13 Issue: 1 Pages: 96-122

    • DOI

      10.5210/dad.2022.104

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Estimating Exchange-Level Annotations with Multitask Learning for Multimodal Dialogue Systems2021

    • Author(s)
      平野 裕貴、岡田 将吾、西本 遥人、駒谷 和範
    • Journal Title

      電子電子情報通信学会論文誌A 基礎・境界

      Volume: J104-A Issue: 2 Pages: 84-94

    • DOI

      10.14923/transfunj.2020HAP0012

    • ISSN
      1881-0195
    • Year and Date
      2021-02-01
    • Related Report
      2020 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Knowledge Graph Completion-based Question Selection for Acquiring Domain Knowledge through Dialogues2021

    • Author(s)
      Komatani Kazunori、Fujioka Yuma、Nakashima Keisuke、Hayashi Katsuhiko、Nakano Mikio
    • Journal Title

      Proc Annual Conference on Intelligent User Interfaces (IUI)

      Volume: - Pages: 531-541

    • DOI

      10.1145/3397481.3450653

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Age Estimation with Speech-Age Model for Heterogeneous Speech Datasets2021

    • Author(s)
      Takeda Ryu、Komatani Kazunori
    • Journal Title

      Proc. Interspeech

      Volume: - Pages: 4164-4168

    • DOI

      10.21437/interspeech.2021-861

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Multimodal Human-Agent Dialogue Corpus with Annotations at Utterance and Dialogue Levels2021

    • Author(s)
      Komatani Kazunori、Okada Shogo
    • Journal Title

      Proc International Conference on Affective Computing & Intelligent Interaction (ACII)

      Volume: - Pages: 1-8

    • DOI

      10.1109/acii52823.2021.9597447

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Multimodal User Satisfaction Recognition for Non-task Oriented Dialogue Systems2021

    • Author(s)
      Wei Wenqing、Li Sixia、Okada Shogo、Komatani Kazunori
    • Journal Title

      Proc. International Conference on Multimodal Interaction (ICMI)

      Volume: - Pages: 586-594

    • DOI

      10.1145/3462244.3479928

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Recognizing Social Signals with Weakly Supervised Multitask Learning for Multimodal Dialogue Systems2021

    • Author(s)
      Hirano Yuki、Okada Shogo、Komatani Kazunori
    • Journal Title

      Proc. International Conference on Multimodal Interaction (ICMI)

      Volume: - Pages: 586-594

    • DOI

      10.1145/3462244.3479927

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Design guidelines for developing systems for dialogue system competitions2021

    • Author(s)
      Kazunori Komatani、Ryu Takeda、Keisuke Nakashima、Mikio Nakano
    • Journal Title

      Proc. International Workshop on Spoken Dialogue Systems (IWSDS)

      Volume: -

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Multiple-Embedding Separation Networks: Sound Class-Specific Feature Extraction for Universal Sound Separation2021

    • Author(s)
      Hokuto Munakata、Ryu Takeda、Kazunori Komatani
    • Journal Title

      Proc. Asia Pacific Signal and Information Processing Association Annual Summit and Conerence (APSIPA ASC)

      Volume: - Pages: 961-967

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Spatial Normalization to Reduce Positional Complexity in Direction-Aided Supervised Binaural Sound Source Separation2021

    • Author(s)
      Ryu Takeda、Kazuhiro Nakadai、Kazunori Komatani
    • Journal Title

      Proc. Asia Pacific Signal and Information Processing Association Annual Summit and Conerence (APSIPA ASC)

      Volume: - Pages: 248-254

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Binarized Embeddings for Fast, Space-Efficient Knowledge Graph Completion2021

    • Author(s)
      Katsuhiko Hayashi、Koki Kishimoto、Masashi Shimbo
    • Journal Title

      IEEE Transactions on Knowledge and Data Engineering

      Volume: -- Pages: 1-13

    • DOI

      10.1109/tkde.2021.3075070

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] User Impressions of Questions to Acquire Lexical Knowledge2020

    • Author(s)
      Kazunori Komatani、Mikio Nakano
    • Journal Title

      Proc. 21st Annual SIGDIAL Meeting on Discourse and Dialogue (SIGDIAL2020)

      Volume: - Pages: 147-156

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Frame-Wise Online Unsupervised Adaptation of DNN-HMM Acoustic Model from Perspective of Robust Adaptive Filtering2020

    • Author(s)
      Takeda Ryu、Komatani Kazunori
    • Journal Title

      Proc. Interspeech

      Volume: - Pages: 1291-1295

    • DOI

      10.21437/interspeech.2020-1301

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] A framework for building closed-domain chat dialogue systems2020

    • Author(s)
      Nakano Mikio、Komatani Kazunori
    • Journal Title

      Knowledge-Based Systems

      Volume: 204 Pages: 106212-106212

    • DOI

      10.1016/j.knosys.2020.106212

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Is She Truly Enjoying the Conversation?2020

    • Author(s)
      Katada Shun、Okada Shogo、Hirano Yuki、Komatani Kazunori
    • Journal Title

      Proceedings of the 2020 International Conference on Multimodal Interaction (ICMI)

      Volume: - Pages: 315-323

    • DOI

      10.1145/3382507.3418844

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Multitask Prediction of Exchange-level Annotations for Multimodal Dialogue Systems2019

    • Author(s)
      Hirano Yuki、Okada Shogo、Nishimoto Haruto、Komatani Kazunori
    • Journal Title

      Proc. International Conference on Multimodal Interaction (ICMI)

      Volume: - Pages: 147-156

    • DOI

      10.1145/3340555.3353730

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Attribute Prediction of Unknown Lexical Entities based on Mixture of Bayesian Segmentation Model2019

    • Author(s)
      Ryu Takeda、Komatani Kazunori
    • Journal Title

      Proc. Life Long Learning for Spoken Language Systems Workshop

      Volume: -

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Presentation] 対話システムにおけるマルチモーダル情報を用いた自己開示の可否判断2022

    • Author(s)
      奥野 尚己, 武田 龍, 駒谷 和範
    • Organizer
      情報処理学会全国大会
    • Related Report
      2021 Annual Research Report
  • [Presentation] ルールと用例を併用してユーザ発話を誘導する対話システムの構築2021

    • Author(s)
      久保祐喜, 生嶋竜実, 二瀬颯斗, 武田龍, 駒谷和範
    • Organizer
      第93回 言語・音声理解と対話処理研究会(第12回対話システムシンポジウム)
    • Related Report
      2021 Annual Research Report
  • [Presentation] システム発話間の整合性を重視した発話選択への深層強化学習の適用2021

    • Author(s)
      黒田佑樹, 武田龍, 駒谷和範
    • Organizer
      第93回 言語・音声理解と対話処理研究会(第12回対話システムシンポジウム)
    • Related Report
      2021 Annual Research Report
  • [Presentation] システム発話間の内容的整合性を用いた強化学習に基づく発話選択2021

    • Author(s)
      黒田 佑樹、武田 龍、駒谷 和範
    • Organizer
      情報処理学会全国大会
    • Related Report
      2020 Annual Research Report
  • [Presentation] グラフデータベース中のエンティティの分布を活用したシステム発話候補の自動抽出2020

    • Author(s)
      中島 圭祐、駒谷 和範
    • Organizer
      第11回対話システムシンポジウム
    • Related Report
      2020 Annual Research Report
  • [Presentation] マルチモーダル対話コーパスHazumi公開と生体信号を含む新規データ収集2020

    • Author(s)
      駒谷 和範、岡田 将吾、堅田 俊
    • Organizer
      第11回対話システムシンポジウム
    • Related Report
      2020 Annual Research Report
  • [Presentation] 対話システムにおける適応的発話選択のためのユーザ状態の設計と推定2020

    • Author(s)
      奥野 尚己, 武田 龍, 駒谷 和範
    • Organizer
      情報処理学会全国大会
    • Related Report
      2019 Annual Research Report
  • [Presentation] User-aware dialogue systems with adaptive responses2019

    • Author(s)
      Kazunori Komatani
    • Organizer
      International Conference on Natural Language Generation (INLG)
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] 対話システムにおける知識獲得質問のためのラベル文字列を用いた知識グラフ補完性能の向上2019

    • Author(s)
      藤岡 勇真, 林 克彦, 中野 幹生, 駒谷 和範
    • Organizer
      人工知能学会第33回全国大会
    • Related Report
      2019 Annual Research Report
  • [Presentation] マルチモーダル情報に基づくユーザの興味度推定のための対話履歴の効果の検証2019

    • Author(s)
      西本 遥人, 武田 龍, 駒谷 和範
    • Organizer
      人工知能学会第33回全国大会
    • Related Report
      2019 Annual Research Report
  • [Presentation] マルチタスク学習を用いたユーザの心象とシステムの話題継続可否の予測モデルの構築2019

    • Author(s)
      平野 裕貴, 岡田 将吾, 西本 遥人, 駒谷 和範
    • Organizer
      電子情報通信学会ヒューマンコミュニケーション基礎(HCS)およびVNV合同研究会 2019年8月研究会
    • Related Report
      2019 Annual Research Report
  • [Presentation] 複数の主観評定を付与した人システム間マルチモーダル対話データの収集と分析2019

    • Author(s)
      駒谷 和範, 岡田 将吾
    • Organizer
      電子情報通信学会ヒューマンコミュニケーション基礎(HCS)およびVNV合同研究会 2019年8月研究会
    • Related Report
      2019 Annual Research Report
  • [Presentation] 未知語属性獲得のための質問の形式がユーザに与える印象の実験的分析2019

    • Author(s)
      駒谷 和範, 中野 幹生
    • Organizer
      第10回対話システムシンポジウム
    • Related Report
      2019 Annual Research Report
  • [Presentation] 雑談対話システム構築フレームワークPyChatに基づく特定シチュエーション向け対話システム2019

    • Author(s)
      中島 圭祐, 駒谷 和範, 中野 幹生
    • Organizer
      第10回対話システムシンポジウム
    • Related Report
      2019 Annual Research Report
  • [Presentation] 対話コーパスに基づく新たなシステム対話行為の設計の検討2019

    • Author(s)
      西本 遥人, 武田 龍, 駒谷 和範
    • Organizer
      第10回対話システムシンポジウム
    • Related Report
      2019 Annual Research Report
  • [Presentation] FoodChatbot: 料理ドメインの雑談対話システム2019

    • Author(s)
      中野 幹生, 駒谷 和範
    • Organizer
      第10回対話システムシンポジウム
    • Related Report
      2019 Annual Research Report
  • [Remarks] 大阪大学 マルチモーダル対話コーパス Hazumi

    • URL

      https://www.nii.ac.jp/dsc/idr/rdata/Hazumi/

    • Related Report
      2020 Annual Research Report
  • [Patent(Industrial Property Rights)] 知識グラフ補完装置、および知識グラフ補完方法2019

    • Inventor(s)
      中野幹生、駒谷和範、林克彦、藤岡勇真
    • Industrial Property Rights Holder
      中野幹生、駒谷和範、林克彦、藤岡勇真
    • Industrial Property Rights Type
      特許
    • Filing Date
      2019
    • Related Report
      2019 Annual Research Report

URL: 

Published: 2019-04-18   Modified: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi