Faculty Profiles - MATSUBARA, Shigeki

Papers - MATSUBARA, Shigeki

Division display 1 - 20 of about 221 ／ All the affair displays >>

Negation Scope Conversion: Towards a Unified Negation-Annotated Dataset Reviewed

Asahi Yoshida, Yoshihide Kato, Shigeki Matsubara

Proceeding of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING2024) 2323.5

On an Intermediate Task for Classifying URL Citations on Scholarly Papers Reviewed

Kazuhiro Wada, Masaya Tsunokake, Shigeki Matsubara

Proceeding of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING2024) 2024.5

語りの傾聴において不同意を示す応答の生成 Reviewed

伊藤滉一朗, 村田匡輝, 大野誠寛, 松原茂樹

自然言語処理 Vol. 31 ( 1 ) 2024.3

Re-Evaluating Syntax-Based Negation Scope Resolution Reviewed

YOSHIDA Asahi, KATO Yoshihide, MATSUBARA Shigeki

IEICE Transactions on Information and Systems Vol. E107.D ( 1 ) page： 165 - 168 2024.1

Authorship：Last author Language：English Publishing type：Research paper (scientific journal) Publisher：The Institute of Electronics, Information and Communication Engineers

<p>Negation scope resolution is the process of detecting the negated part of a sentence. Unlike the syntax-based approach employed in previous researches, state-of-the-art methods performed better without the explicit use of syntactic structure. This work revisits the syntax-based approach and re-evaluates the effectiveness of syntactic structure in negation scope resolution. We replace the parser utilized in the prior works with state-of-the-art parsers and modify the syntax-based heuristic rules. The experimental results demonstrate that the simple modifications enhance the performance of the prior syntax-based method to the same level as state-of-the-art end-to-end neural-based methods.</p>

DOI： 10.1587/transinf.2023edl8050

Research data storage URL： https://nagoya.repo.nii.ac.jp/records/2007004

Web of Science

Scopus

CiNii Research

Identifying Influential References in Scholarly Papers Using Citation Contexts Reviewed

Tomoki Ikoma, Shigeki Matsubara

Proceedings of the 25th International Conference on Asia-Pacific Digital Libraries (ICADL-2023) page： 152 - 161 2023.12

Paper Recommendation Using Citation Contexts in Scholarly Documents Reviewed

Tomoki Ikoma, Shigeki Matsubara

Proceedings of the 37th Pacific Asia Conference on Language, Information and Computation (PACLIC 37) 2023.12

Automatic Insertion of Commas and Linefeeds into Lecture Transcripts Based on Multi-task Learning Reviewed

Zhicheng Fang, Masaki Murata, Shigeki Matsubara:

Proceedings of the 37th Pacific Asia Conference on Language, Information and Computation (PACLIC 37) 2023.12

Identifying Influential References in Scholarly Papers Using Citation Contexts Reviewed

Ikoma T., Matsubara S.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 14458 page： 152 - 161 2023.12

On the Use of Language Models for Function Identification of Citations in Scholarly Papers Reviewed

Tomoki Ikoma and Shigeki Matsubara

Proceedings of the 2nd Workshop on Information Extraction from Scientific Publications (WIESP 2023) 2023.11

Revisiting Syntax-Based Approach in Negation Scope Resolution Reviewed

Asahi Yoshida, Yoshihide Kato and Shigeki Matsubara

Proceedings of the 12th Joint Conference on Lexical and Computational Semantics (*SEM2023) page： 18 - 23 2023.7

Natural Language Inference Using Neural Network and Tableau Method Reviewed

Saji Ayahito, Takao Daiki, Kato Yoshihide, Matsubara Shigeki

Journal of Natural Language Processing Vol. 30 ( 2 ) page： 632 - 663 2023.6

Authorship：Corresponding author Language：Japanese Publishing type：Research paper (scientific journal) Publisher：The Association for Natural Language Processing

<p>Natural language inference (NLI) is the task of identifying the inferential relation between a text pair. In recent times, neural-based approaches have achieved high performance in NLI. However, they are unable to explain their reasoning processes. On the other hand, symbolic approaches have the advantage that their reasoning process is understandable to humans. This paper proposes a method for integrating a neural NLI model and the tableau proof system, with the latter explaining the reasoning processes. The standard tableau method decomposes logical formulas by applying inferential rules and checks for a valuation that satisfies the given constraints. Unlike the standard tableau method, the proposed method uses dependency structures as its components rather than logical formulas and employs a neural NLI model for the latter process. To analyze the behavior of our method, we conducted an experiment on the neural NLI model and the proposed method using SNLI corpus. In addition, we formalize our method model-theoretically and clarify the theoretical limitations of this method based on the formalization. </p>

DOI： 10.5715/jnlp.30.632

Research data storage URL： https://nagoya.repo.nii.ac.jp/records/2003368

CiNii Research

Generation Timing Estimation of Responsive Utterances to Show Attentive Listening to Narrative Speech Reviewed

ITO Koichiro, MURATA Masaki, OHNO Tomohiro, MATSUBARA Shigeki

Vol. J106-A ( 3 ) page： 136 - 145 2023.3

Authorship：Last author Language：Japanese Publishing type：Research paper (scientific journal) Publisher：The Institute of Electronics, Information and Communication Engineers

Conversational agents, e.g., communication robots and smart speakers, are expected to listen to human narratives. For realizing such agents, it is necessary to generate responsive utterances at appropriate timings. This paper proposes a method to estimate the degree of appropriateness as generation timings of responsive utterances. The proposed method estimates the degree by encoding acoustic and linguistic narrative information with a transformer-based approach. To evaluate the performance, an experiment was conducted using 131,616 responsive utterances. The experimental results demonstrated that the proposed method was effective. Additionally, we confirmed that the degree of appropriateness was useful for response timing detection.

DOI： 10.14923/transfunj.2022hap0013

CiNii Research

Bottom-up Japanese Word Ordering using BERT Reviewed

Masato Yamazoe, Tomohiro Ohno and Shigeki Matsubara

Proceedings of the 15th International Conference on Agents and Artificial Intelligence (ICAART-2023) Vol. 3 page： 673 - 681 2023.2

Japanese Word Reordering based on Topological Sort Reviewed

Peng Sun, Tomohiro Ohno and Shigeki Matsubara

Proceedings of the 15th International Conference on Agents and Artificial Intelligence (ICAART-2023) Vol. 3 page： 768 - 775 2023.2

Question Difficulty Decision for Automated Interview in Foreign Language Test Reviewed

Minghao Lin, Koichiro Ito and Shigeki Matsubara

Proceedings of the 15th International Conference on Agents and Artificial Intelligence (ICAART-2023) Vol. 3 page： 768 - 775 2023.2

Japanese Word Reordering Based on Topological Sort

Sun P., Ohno T., Matsubara S.

International Conference on Agents and Artificial Intelligence Vol. 3 page： 768 - 775 2023

Language：English Publishing type：Research paper (scientific journal) Publisher：International Conference on Agents and Artificial Intelligence

In Japanese, some sentences are grammatically well-formed, but not easy to read. This paper proposes a method for Japanese word reordering, which first adapts a Japanese BERT model to predict suitable pairwise orderings between two words in a sentence, and then converts the predicted results into a graph. The vertices in the graph represent the words in the sentence. The edges represent the pairwise orderings between two words. Finally, topological sort is applied to find the correct word ordering in a sentence by visiting each graph vertex. We conducted an evaluation experiment with uneasy-to-read Japanese sentences created from newspaper article sentences.

DOI： 10.5220/0011775000003393

Scopus

Question Difficulty Decision for Automated Interview in Foreign Language Test

Lin M., Ito K., Matsubara S.

International Conference on Agents and Artificial Intelligence Vol. 3 page： 760 - 767 2023

Language：English Publishing type：Research paper (scientific journal) Publisher：International Conference on Agents and Artificial Intelligence

As globalization continues to progress all over the world, demand is growing for objective and rapid assessment of language proficiency in foreign language learners. While automated assessment of listening, reading, and writing skills has been proposed, little research has been done to automate assessment of speaking skills. In this paper, we propose a method of deciding the difficulty of questions generated for interview tests of skills assessment in speaking a foreign language. To address question difficulty flexibly according to the abilities of test takers, our method considers the appropriateness of responses from test takers. We implemented this method using the large-scale pre-trained language model BERT (Bidirectional Encoder Representation from Transformers). Experiments were conducted using simulated test data from the Japanese Learner’s Conversation Database to confirm the effectiveness of our method in deciding difficulty.

DOI： 10.5220/0011774900003393

Scopus

Extracting Information about Research Resources from Scholarly Papers Reviewed

Ayahito Saji and Shigeki Matsubara

Proceedings of the 24th International Conference on Asia-Pacific Digital Libraries (ICADL-2022) 2022.12

Classification of URL Citations in Scholarly Papers for Promoting Utilization of Research Artifacts Reviewed

Masaya Tsunokake and Shigeki Matsubara

Proceedings of AACL-IJCNLP-2022 The 1st Workshop on Information Extraction from Scientific Publications 2022.11

A Model-Theoretic Formalization of Natural Language Inference Using Neural Network and Tableau Method Reviewed

Ayahito Saji, Yoshihide Kato, Shigeki Matsubara

Proceedings of the 36th Pacific Asia Conference on Language, Information and Computation (PACLIC 35) 2022.10

　 PREV - NEXT 　

<MATSUBARA, Shigeki>

Papers - MATSUBARA, Shigeki