Updated on 2024/04/24

写真a

 
Makoto Miwa
 
Organization
Graduate School of Engineering Department of Advanced Science and Technology Electronics and Information Engineering Knowledge and Data Engineering Professor   
Degree
Phd (Science) （ 2008.3   The University of Tokyo ）
External link
Contact information
メールアドレス

Research Areas

  • Informatics / Intelligent informatics

  • Informatics / Entertainment and game informatics

Main research papers

  • End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures

    Makoto Miwa and Mohit Bansal,

    Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics   ( 1 )   1105 - 1116   2016.8.8

    Modeling Joint Entity and Relation Extraction with Table Representation

    Makoto Miwa ,Yutaka Sasaki

    Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing   1858 - 1869   2014.10.25

    A method for integrating and ranking the evidence for biochemical pathways by mining reactions from text

    Makoto Miwa, Tomoko Ohta, Rafal Rak, Andrew Rowley, Douglas B. Kell, Sampo Pyysalo and Sophia Ananiadou

    Bioinformatics   29 ( 13 )   i44 - i52   2013.7.1

    Boosting automatic event extraction from the literature using domain adaptation and coreference resolution

    Makoto Miwa, Paul Thompson and Sophia Ananiadou

    Bioinformatics   28 ( 13 )   1759 - 1765   2012.7.1

    A Rich Feature Vector for Protein-Protein Interaction Extraction from Multiple Corpora

    Makoto Miwa, Rune Sætre, Yusuke Miyao, and Jun'ichi Tsujii,

    Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing   121 - 130   2009.8.6

Research History

  • Toyota Technological Institute   Graduate School of Engineering Department of Advanced Science and Technology Electronics and Information Engineering Knowledge and Data Engineering   Professor

    2024

  • Toyota Technological Institute   Graduate School of Engineering   Associate Professor

    2014 - 2023

  • The University of Manchester   School of Computer Science   Research Associate

    2011 - 2014

  • The University of Tokyo   Graduate School of Infomation Science and Technology   Researcher

    2008 - 2011

▼display all

Education

  • The University of Tokyo   Graduate School of Frontier Sciences   Department of Frontier Informatics

    2003.4 - 2008.3

    　 More details

    Country： Japan

Professional Memberships

  • 情報処理学会

  • 言語処理学会

  • 電子情報通信学会

  • 人工知能学会

Research theme

  • Knowledge-based dialogue generation

    佐々木　裕, 三輪　誠

    2017

  • Machine Learning for driving a small UMV

    佐々木　裕, 三輪　誠

    2017

  • Natural Language Processing with Deep Learning

    三輪　誠, 佐々木　裕

    2017

  • Well-formedness conditions of Japanese Sign Language syllables in terms of combinations of syllable constituents

    原　大介, 三輪　誠

    2017

  • Structural Relation Extraction from Natural Language Text

    三輪　誠, 佐々木　裕

    2017

  • Constructing Driving Knowledge Bases

    佐々木　裕, 三輪　誠

    2017

  • 知識グラフ上での表現学習

    三輪　誠, 佐々木　裕

    2017

    　More details

    様々な分野の知識をノードと関係で表した知識グラフが盛んに開発・利用されている。本研究では、様々なデータ型や種類のノードを含むマルチモーダルな情報やトリプルを超えた複数の関係に焦点を当て、従来の単一種類のノードを用いたトリプル上での表現学習を超えた新しい表現学習手法について研究を進めている。

    Outcome：

    2023
    知識グラフ内の1対多の関係を明示的に考慮し、学習を行う手法を提案した。

    2022
    薬物を対象に、薬物の説明文、構造、関係するタンパク質などを考慮した知識グラフを対象に、表現学習を行い、知識グラフの欠損部分を補完する手法を考案した。

    2017
    書誌情報を対象として、タイトルや概要などのテキストと、著者・出版年などを考慮した知識グラフを構築し、それぞれの要素における表現学習を行い、得られた表現間の距離を測ることでソフトな検索を行う、書士検索システムを提案した。

  • Applying AI methods to Machining, Forge Design, and CAE

    佐々木　裕, 三輪　誠, 古谷　克司

    2017

  • Systematic Review with Machine Learning

    佐々木　裕, 三輪　誠

    2017 - 2021

  • Research on Word Embedding Vecotrs

    佐々木　裕, 三輪　誠

    2017 - 2021

▼display all

Papers

  • Two evaluations on Ontology-style relation annotations Reviewed

    Savong Bou, Makoto Miwa, Yutaka Sasaki

    Computer Speech and Language   84   2024.3 （   ISSN:0885-2308   eISSN:1095-8363 ）

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    In this paper, we propose an Ontology-Style Relation (OSR) annotation approach. In conventional Relation Extraction (RE) datasets, relations are annotated as a link between two entity mentions. In contrast, in our OSR annotation, a relation is annotated as a relation mention (i.e., not a link but a node) and rdfs:domain and rdfs:range links are annotated from the relation mention to its argument entity mentions. This approach has the following benefits: (1) the relation annotations can be easily converted to Resource Description Framework (RDF) triples to populate an Ontology, (2) some part of conventional RE tasks can be tackled as Named Entity Recognition (NER) tasks, and the relation classes are limited to several RDF properties, and (3) OSR annotations can be used for clear documentations of Ontology contents. We conducted two kinds of evaluation to investigate effects of OSR annotation. We converted (1) an in-house corpus of Japanese Rules of the Road (RoR) in conventional annotations into the OSR annotations and built a novel OSR-RoR corpus and (2) SemEval-2010 Task 8 dataset into the OSR annotations (called OSR-SemEval corpus). We compared the NER and RE performance using neural NER/RE tool DyGIE++ on the conventional and OSR annotations. The experimental results show that the OSR annotations make the RE task easier while introducing slight complexity into the NER task.

    DOI： 10.1016/j.csl.2023.101569

  • Contextualized medication event extraction with striding NER and multi-turn QA. Reviewed

    Tomoki Tsujimura, Koshi Yamada, Ryuki Ida, Makoto Miwa, Yutaka Sasaki

    J. Biomed. Informatics   144   104416 - 104416   2023.8 （   ISSN:1532-0464 ）

    　More details

    Authorship：Corresponding author   Language：English   Publishing type：Research paper (scientific journal)  

    This paper describes contextualized medication event extraction for automatically identifying medication change events with their contexts from clinical notes. The striding named entity recognition (NER) model extracts medication name spans from an input text sequence using a sliding-window approach. Specifically, the striding NER model separates the input sequence into a set of overlapping subsequences of 512 tokens with 128 tokens of stride, processing each subsequence using a large pre-trained language model and aggregating the outputs from the subsequences. The event and context classification has been done with multi-turn question-answering (QA) and span-based models. The span-based model classifies the span of each medication name using the span representation of the language model. In the QA model, event classification is augmented with questions in classifying the change events of each medication name and the context of the change events, while the model architecture is a classification style that is the same as the span-based model. We evaluated our extraction system on the n2c2 2022 Track 1 dataset, which is annotated for medication extraction (ME), event classification (EC), and context classification (CC) from clinical notes. Our system is a pipeline of the striding NER model for ME and the ensemble of the span-based and QA-based models for EC and CC. Our system achieved a combined F-score of 66.47% for the end-to-end contextualized medication event extraction (Release 1), which is the highest score among the participants of the n2c2 2022 Track 1.

    DOI： 10.1016/j.jbi.2023.104416

  • Integrating heterogeneous knowledge graphs into drug-drug interaction extraction from the literature Reviewed

    Masaki Asada, Makoto Miwa, Yutaka Sasaki

    Bioinformatics (Oxford, England)   39 ( 1 )   2023.1 （     eISSN:1367-4811 ）

    　More details

    Authorship：Corresponding author   Language：English   Publishing type：Research paper (scientific journal)  

    MOTIVATION: Most of the conventional deep neural network-based methods for drug-drug interaction (DDI) extraction consider only context information around drug mentions in the text. However, human experts use heterogeneous background knowledge about drugs to comprehend pharmaceutical papers and extract relationships between drugs. Therefore, we propose a novel method that simultaneously considers various heterogeneous information for DDI extraction from the literature. RESULTS: We first construct drug representations by conducting the link prediction task on a heterogeneous pharmaceutical knowledge graph (KG) dataset. We then effectively combine the text information of input sentences in the corpus and the information on drugs in the heterogeneous KG (HKG) dataset. Finally, we evaluate our DDI extraction method on the DDIExtraction-2013 shared task dataset. In the experiment, integrating heterogeneous drug information significantly improves the DDI extraction performance, and we achieved an F-score of 85.40%, which results in state-of-the-art performance. We evaluated our method on the DrugProt dataset and improved the performance significantly, achieving an F-score of 77.9%. Further analysis showed that each type of node in the HKG contributes to the performance improvement of DDI extraction, indicating the importance of considering multiple pieces of information. AVAILABILITY AND IMPLEMENTATION: Our code is available at https://github.com/tticoin/HKG-DDIE.git.

    DOI： 10.1093/bioinformatics/btac754

  • Biomedical Document Classification with Literature Graph Representations of Bibliographies and Entities. Reviewed

    Ryuki Ida, Makoto Miwa, Yutaka Sasaki

    BioNLP@ACL   385 - 395   2023 （ ISBN:9781959429852   ISSN:0736-587X ）

    　More details

    Language：English   Publishing type：Research paper (international conference proceedings)  

    This paper proposes a new document classification method that incorporates the representations of a literature graph created from bibliographic and entity information. Recently, document classification performance has been significantly improved with large pre-trained language models; however, there still remain documents that are difficult to classify. External information, such as bibliographic information, citation links, descriptions of entities, and medical taxonomies, has been considered one of the keys to dealing with such documents in document classification. Although several document classification methods using external information have been proposed, they only consider limited relationships, e.g., word co-occurrence and citation relationships. However, there are multiple types of external information. To overcome the limitation of the conventional use of external information, we propose a document classification model that simultaneously considers bibliographic and entity information to deeply model the relationships among documents using the representations of the literature graph. The experimental results show that our proposed method outperforms existing methods on two document classification datasets in the biomedical domain with the help of the literature graph. Our source code is publicly available at https://github.com/tticoin/BDCL-LitGraph.

    DOI： 10.18653/v1/2023.bionlp-1.36

    Other Link： https://dblp.uni-trier.de/rec/conf/bionlp/2023

  • Biomedical Relation Extraction with Entity Type Markers and Relation-specific Question Answering. Reviewed

    Koshi Yamada, Makoto Miwa, Yutaka Sasaki

    BioNLP@ACL   377 - 384   2023 （ ISBN:9781959429852   ISSN:0736-587X ）

    　More details

    Language：English   Publishing type：Research paper (international conference proceedings)  

    Recently, several methods have tackled the relation extraction task with QA and have shown successful results. However, the effectiveness of existing methods in specific domains, such as the biomedical domain, is yet to be verified. When there are multiple entity pairs that share an entity in a sentence, a QA-based relation extraction model that outputs only one single answer to a given question may not extract desired relations. In addition, these methods employ QA models that are not tuned for relation extraction. To address these issues, we first extend and apply a span QA-based relation extraction method to the drug-protein relation extraction by creating question templates and incorporating entity type markers. We further propose a binary QA-based method that directly uses the entity information available in the relation extraction task. The experimental results on the DrugProt dataset show that our QA-based methods, especially the proposed binary QA method, are effective for drug-protein relation extraction. Our source code is available at https://github.com/tticoin/BioRE-ETM-QA.

    DOI： 10.18653/v1/2023.bionlp-1.35

    Other Link： https://dblp.uni-trier.de/rec/conf/bionlp/2023

  • Distantly Supervised Document-Level Biomedical Relation Extraction with Neighborhood Knowledge Graphs. Reviewed

    Takuma Matsubara, Makoto Miwa, Yutaka Sasaki

    BioNLP@ACL   363 - 368   2023 （ ISBN:9781959429852   ISSN:0736-587X ）

    　More details

    Language：English   Publishing type：Research paper (international conference proceedings)  

    We propose a novel distantly supervised document-level biomedical relation extraction model that uses partial knowledge graphs that include the graph neighborhood of the entities appearing in each input document. Most conventional distantly supervised relation extraction methods use only the entity relations automatically annotated by using knowledge base entries. They do not fully utilize the rich information in the knowledge base, such as entities other than the target entities and the network of heterogeneous entities defined in the knowledge base. To address this issue, our model integrates the representations of the entities acquired from the neighborhood knowledge graphs with the representations of the input document. We conducted experiments on the ChemDisGene dataset using Comparative Toxicogenomics Database (CTD) for document-level relation extraction with respect to interactions between drugs, diseases, and genes. Experimental results confirmed the performance improvement by integrating entities and their neighborhood biochemical information from the knowledge base.

    DOI： 10.18653/v1/2023.bionlp-1.33

    Other Link： https://dblp.uni-trier.de/rec/conf/bionlp/2023

  • Large-scale neural biomedical entity linking with layer overwriting. Reviewed

    Tomoki Tsujimura, Makoto Miwa, Yutaka Sasaki

    J. Biomed. Informatics   143   104433 - 104433   2023 （   ISSN:1532-0464 ）

    　More details

    Authorship：Corresponding author   Language：English   Publishing type：Research paper (scientific journal)  

    Motivation: Entity linking is the task of linking entity mentions to the database entries corresponding to the entity mentions. Entity linking enables the treatment of superficially different but semantically identical mentions as the same entity. Since millions of concepts are listed in biomedical databases, selecting the correct database entry for each targeted entity is challenging. Simple string matching between the word and each synonym in biomedical databases is insufficient to handle a wide variety of variants of biomedical entities appearing in the biomedical literature. Recent progress in neural approaches is promising for entity linking. Still, existing neural methods require sufficient data, which is difficult to prepare in biomedical entity linking that deals with millions of biomedical concepts. Therefore, we need to develop a new neural method to train entity-linking models over the sparse training data covering a very limited part of the biomedical concepts. Results: We have devised a pure neural model that classifies biomedical entity mentions into millions of biomedical concepts. The classifier employs (1) the layer overwriting that breaks through the performance ceiling during training, (2) training data augmentation using database entries that compensate for the problem of insufficient training data, and (3) the cosine similarity-based loss function that helps distinguish the millions of biomedical concepts. Our system using the proposed classifier was ranked first in the official run of the National NLP Clinical Challenges (n2c2) 2019 Track 3, which targeted linking medical/clinical entity mentions to 434,056 Concept Unique Identifier (CUI) entries. We also applied our system to the MedMentions dataset, which has 3.2M candidate concepts. Experimental results confirmed the same advantages of our proposed method. We further evaluated our system on the NLM-CHEM corpus with 350K candidate concepts, and our system achieved a new state-of-the-art performance on the corpus. Availability: https://github.com/tti-coin/bio-linking Contact: makoto.miwa@toyota-ti.ac.jp

    DOI： 10.1016/j.jbi.2023.104433

  • Editing Relation Candidate Edges of Relation Graphs for Document-Level Relation Extraction Reviewed

    牧野晃平, 三輪誠, 佐々木裕

    自然言語処理(Web)   30 ( 2 )   2023 （   ISSN:2185-8314 ）

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)  

  • Integrating heterogeneous knowledge graphs into drug–drug interaction extraction from the literature Reviewed

    Masaki Asada*, Makoto Miwa, Yutaka Sasaki

    Bioinformatics   39 ( 1 )   btac754   2022.11 （   ISSN:1367-4811 ）

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：Oxford University Press  

  • Comparing neural models for nested and overlapping biomedical event detection Reviewed

    Kurt Espinosa **, Panagiotis Georgiadis **, Fenia Christopoulou **, Meizhi Ju **, Makoto Miwa, Sophia Ananiadou **

    BMC Bioinformatics   23 ( 1 )   211   2022.6 （   ISSN:1471-2105 ）

    　More details

    Publishing type：Research paper (scientific journal)  

  • Improving Supervised Drug-Protein Relation Extraction with Distantly Supervised Models Reviewed

    Naoki Iinuma*, Makoto Miwa, Yutaka Sasaki

    The Proceedings of the 2022 Workshop on Biomedical Natural Language Processing   161 - 170   2022.5

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • Extracting and Analyzing Inorganic Material Synthesis Procedures in the Literature. Reviewed

    Kohei Makino, Fusataka Kuniyoshi, Jun Ozawa, Makoto Miwa

    IEEE Access   10   31524 - 31537   2022

    　More details

    Authorship：Last author   Language：English   Publishing type：Research paper (scientific journal)  

    DOI： 10.1109/ACCESS.2022.3160201

  • BioVAE: a pre-trained latent variable language model for biomedical text mining. Reviewed International coauthorship

    Hai-Long Trieu, Makoto Miwa, Sophia Ananiadou

    Bioinform.   38 ( 3 )   872 - 874   2022

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    DOI： 10.1093/bioinformatics/btab702

  • 日本手話における手型変化 Reviewed

    原　大介, 三輪　誠

    日本手話学会第47回大会予稿集   10 - 11   2021.12

    　More details

    Authorship：Last author   Publishing type：Research paper (international conference proceedings)   Publisher：日本手話学会  

  • TTI-COIN at BioCreative VII Track 2 Reviewed

    Tomoki Tsujimura*, Ryuki Ida*, Makoto Miwa, Yutaka Sasaki

    BioCreative VII Workshop   2021.11

    　More details

    Publishing type：Research paper (international conference proceedings)  

  • TTI-COIN at BioCreative VII Track 1 Reviewed

    Naoki Iinuma*, Masaki Asada*, Makoto Miwa, Yutaka Sasaki

    BioCreative VII Workshop   2021.11

    　More details

    Publishing type：Research paper (international conference proceedings)  

  • A Neural Edge-Editing Approach for Document-Level Relation Graph Extraction Reviewed

    Kohei Makino*, Makoto Miwa, Yutaka Sasaki

    Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021   2021.8

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • Representing a Heterogeneous Pharmaceutical Knowledge-Graph with Textual Information Reviewed

    Masaki Asada*, NALLAPPAN, Gunasekaran, Makoto Miwa, Yutaka Sasaki

    Front. Res. Metr. Anal.   6   2021.7

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：Frontiers  

  • Analyzing Research Trends in Inorganic Materials Literature Using NLP. Reviewed

    Fusataka Kuniyoshi, Jun Ozawa, Makoto Miwa

    Machine Learning and Knowledge Discovery in Databases. Applied Data Science Track - European Conference   319 - 334   2021 （ ISBN:9783030865160 ）

    　More details

    Authorship：Last author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Springer  

    DOI： 10.1007/978-3-030-86517-7_20

    Other Link： https://dblp.uni-trier.de/db/conf/pkdd/pkdd2021-5.html#KuniyoshiOM21

  • Distantly Supervised Relation Extraction with Sentence Reconstruction and Knowledge Base Priors. Reviewed International coauthorship

    Fenia Christopoulou, Makoto Miwa, Sophia Ananiadou

    Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies(NAACL-HLT)   11 - 26   2021 （ ISBN:9781954085466 ）

    　More details

    Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

    DOI： 10.18653/v1/2021.naacl-main.2

    Other Link： https://dblp.uni-trier.de/rec/conf/naacl/2021

  • DeepEventMine: end-to-end neural nested event extraction from biomedical texts. Reviewed International coauthorship

    Hai-Long Trieu, Thy Thy Tran, Khoa N A Duong, Anh Nguyen, Makoto Miwa, Sophia Ananiadou

    Bioinformatics (Oxford, England)   36 ( 19 )   4910 - 4917   2020.12

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    MOTIVATION: Recent neural approaches on event extraction from text mainly focus on flat events in general domain, while there are less attempts to detect nested and overlapping events. These existing systems are built on given entities and they depend on external syntactic tools. RESULTS: We propose an end-to-end neural nested event extraction model named DeepEventMine that extracts multiple overlapping directed acyclic graph structures from a raw sentence. On the top of the bidirectional encoder representations from transformers model, our model detects nested entities and triggers, roles, nested events and their modifications in an end-to-end manner without any syntactic tools. Our DeepEventMine model achieves the new state-of-the-art performance on seven biomedical nested event extraction tasks. Even when gold entities are unavailable, our model can detect events from raw text with promising performance. AVAILABILITY AND IMPLEMENTATION: Our codes and models to reproduce the results are available at: https://github.com/aistairc/DeepEventMine. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

    DOI： 10.1093/bioinformatics/btaa540

  • 再帰ニューラルネットを用いた車両運動性の代理モデリング Reviewed

    牧野晃平 知能数理*, 新谷浩平*, 阿部充治*, 三輪　誠, 佐々木　裕

    日本機械学会論文集   86 ( 891 )   20 - 00177   2020.11 （   ISSN:2187-9761 ）

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：一般社団法人 日本機械学会  

  • mgsohrab at WNUT 2020 Shared Task-1: Neural Exhaustive Approach for Entity and Relation Recognition Over Wet Lab Protocols Reviewed

    Mohammad Golam Sohrab **, Anh-Khoa Duong Nguyen **, Makoto Miwa, Hiroya Takamura **

    Proceedings of the Sixth Workshop on Noisy User-generated Text (W-NUT 2020)   290 - 298   2020.11

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • Syntactically-Informed Word Representations from Graph Neural Network Reviewed

    Thy Thy Tran **, Makoto Miwa, Sophia Ananiadou **

    Neurocomputing   413 ( 6 )   431 - 443   2020.11 （   ISSN:0925-2312 ）

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：Elsevier  

  • Using Drug Descriptions and Molecular Structures for Drug-Drug Interaction Extraction from Literature Reviewed

    浅田真生 知能数理*, Makoto Miwa, Yutaka Sasaki

    Bioinformatics   2020.10 （   ISSN:1367-4803 ）

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：Oxford University Press  

  • BENNERD: A Neural Named Entity Linking System for COVID-19 Reviewed

    Mohammad Golam Sohrab **, Khoa Duong **, Makoto Miwa, Goran Topic **, Ikeda Masami **, Hiroya Takamura **

    Proceedings of 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP 2020): System Demonstrations   182 - 188   2020.10

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • DeepEventMine: : End-to-end Neural Nested Event Extraction from Biomedical Texts Reviewed

    Hai-Long Trieu **, Thy Thy Tran **, Khoa N. A. Duong **, Anh Nguyen **, Makoto Miwa, Sophia Ananiadou **

    Bioinformatics   36 ( 19 )   4910 - 4917   2020.10 （   ISSN:1367-4803 ）

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：Oxford University Press  

  • BENNERD: A Neural Named Entity Linking System for COVID-19 Reviewed

    Mohammad Golam Sohrab **, Khoa Duong **, Makoto Miwa, Goran Topic **, Ikeda Masami **, Hiroya Takamura **

    Proceedings of 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP 2020): System Demonstrations   182 - 188   2020.10

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • 関係分類における依存木上の重要トークンの自動判別 Reviewed

    辻村有輝 知能数理*, 三輪　誠, 佐々木　裕

    Journal of Natural Language Processing   27 ( 2 )   211 - 235   2020.6 （   ISSN:1340-7619 ）

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：言語処理学会  

  • Ontology-Style Relation Annotation: A Case Study Reviewed

    Savong Bou 知能数理*, Makoto Miwa, Yutaka Sasaki

    12th Language Resources and Evaluation Conference (LREC-2020)   4867 - 4876   2020.5

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：European Language Resources Association  

  • コーディングと動画を併用した日本手話音節の適格性予測 Reviewed

    高藤朋史*, 三輪　誠, 原　大介

    言語処理学会第26回年次大会発表論文集   259 - 262   2020.3

    　More details

    Publishing type：Research paper (international conference proceedings)  

  • Ontology-Style Relation Annotation: A Case Study Reviewed

    サーヴォン　ブー, Naoki Suzuki 知能数理*, Makoto Miwa, Yutaka Sasaki

    12th Edition of Language Resources and Evaluation Conference   2020.3

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Ontology-Style Relation Annotation: A Case Study (accepted)  

  • Adverse Drug Events and Medication Relation Extraction in EHRs with Ensemble Deep Learning Methods Reviewed

    Fenia Christopoulou **, Thy Thy Tran **, Sunil Kumar Sahu **, Makoto Miwa, Sophia Ananiadou **

    Journal of the American Medical Informatics Association   27 ( 1 )   39 - 46   2020.1

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：American Medical Informatics Association  

  • An Ensemble of Neural Models for Nested Adverse Drug Events and Medication Extraction with Subwords Reviewed

    Meizhi Ju **, Nhung T.H. Nguyen **, Makoto Miwa, Sophia Ananiadou **

    Journal of the American Medical Informatics Association   27 ( 1 )   22 - 30   2020.1

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：American Medical Informatics Association  

  • Adverse drug events and medication relation extraction in electronic health records with ensemble deep learning methods. Reviewed International coauthorship

    Fenia Christopoulou, Thy Thy Tran, Sunil Kumar Sahu, Makoto Miwa, Sophia Ananiadou

    J. Am. Medical Informatics Assoc.   27 ( 1 )   39 - 46   2020 （   ISSN:1067-5027   eISSN:1527-974X ）

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    OBJECTIVE: Identification of drugs, associated medication entities, and interactions among them are crucial to prevent unwanted effects of drug therapy, known as adverse drug events. This article describes our participation to the n2c2 shared-task in extracting relations between medication-related entities in electronic health records. MATERIALS AND METHODS: We proposed an ensemble approach for relation extraction and classification between drugs and medication-related entities. We incorporated state-of-the-art named-entity recognition (NER) models based on bidirectional long short-term memory (BiLSTM) networks and conditional random fields (CRF) for end-to-end extraction. We additionally developed separate models for intra- and inter-sentence relation extraction and combined them using an ensemble method. The intra-sentence models rely on bidirectional long short-term memory networks and attention mechanisms and are able to capture dependencies between multiple related pairs in the same sentence. For the inter-sentence relations, we adopted a neural architecture that utilizes the Transformer network to improve performance in longer sequences. RESULTS: Our team ranked third with a micro-averaged F1 score of 94.72% and 87.65% for relation and end-to-end relation extraction, respectively (Tracks 2 and 3). Our ensemble effectively takes advantages from our proposed models. Analysis of the reported results indicated that our proposed approach is more generalizable than the top-performing system, which employs additional training data- and corpus-driven processing techniques. CONCLUSIONS: We proposed a relation extraction system to identify relations between drugs and medication-related entities. The proposed approach is independent of external syntactic tools. Analysis showed that by using latent Drug-Drug interactions we were able to significantly improve the performance of non-Drug-Drug pairs in EHRs.

    DOI： 10.1093/jamia/ocz101

    Other Link： https://dblp.uni-trier.de/db/journals/jamia/jamia27.html#ChristopoulouTS20

  • Annotating and Extracting Synthesis Process of All-Solid-State Batteries from Scientific Literature. Reviewed

    Fusataka Kuniyoshi, Kohei Makino, Jun Ozawa, Makoto Miwa

    Proceedings of The 12th Language Resources and Evaluation Conference(LREC)   1941 - 1950   2020

    　More details

    Authorship：Last author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：European Language Resources Association  

    Other Link： https://dblp.uni-trier.de/conf/lrec/2020

  • Connecting the Dots: Document-level Neural Relation Extraction with Edge-oriented Graphs Reviewed

    Fenia Christopoulou **, Makoto Miwa, Sophia Ananiadou **

    Proceedings of 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP 2019)   4927 - 4938   2019.11

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • A Search-based Neural Model for Biomedical Nested and Overlapping Event Detection Reviewed

    Kurt Espinosa **, Makoto Miwa, Sophia Ananiadou **

    Proceedings of 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP 2019)   3670 - 3677   2019.11

    　More details

    Publishing type：Research paper (international conference proceedings)  

  • Coreference Resolution in Full Text Articles with BERT and Syntax-based Mention Filtering Reviewed

    Hai-Long Trieu **, Anh-Khoa Duong Nguyen **, Nhung Nguyen **, Makoto Miwa, Hiroya Takamura **, Sophia Ananiadou **

    Proceedings of The 5th Workshop on BioNLP Open Shared Tasks   196 - 205   2019.11

    　More details

    Publishing type：Research paper (international conference proceedings)  

  • A Neural Pipeline Approach for the PharmaCoNER Shared Task using Contextual Exhaustive Models Reviewed

    Mohammad Golam Sohrab **, Pham Minh Thang **, Makoto Miwa, Hiroya Takamura **

    Proceedings of The 5th Workshop on BioNLP Open Shared Tasks   47 - 55   2019.11

    　More details

    Publishing type：Research paper (international conference proceedings)  

  • A Generic Neural Exhaustive Approach for Entity Recognition and Sensitive Span Detection Reviewed

    Mohammad Golam Sohrab **, Pham Minh Thang **, Makoto Miwa

    Proceedings of the Iberian Languages Evaluation Forum (IberLEF 2019)   735 - 743   2019.9

    　More details

    Authorship：Last author   Publishing type：Research paper (international conference proceedings)  

  • Inter-sentence Relation Extraction with Document-level Graph Convolutional Neural Network Reviewed

    Sunil Kumar Sahu **, Fenia Christopoulou **, Makoto Miwa, Sophia Ananiadou **

    Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL 2019)   4309 - 4316   2019.7

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • Inter-sentence Relation Extraction with Document-level Graph Convolutional Neural Network Reviewed

    Sunil Kumar Sahu The University of Manchester*, Fenia Christopoulou The University of Manchester*, Makoto Miwa, Sophia Ananiadou The University of Manchester*

    Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL 2019)   4309 - 4316   2019.7

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • 深層学習を用いた日本手話音節の適格性解析 Reviewed

    高藤朋史 知能数理*, 三輪　誠, 佐々木　裕, 原　大介

    言語処理学会第25回年次大会発表論文集   486 - 489   2019.3

    　More details

    Publishing type：Research paper (international conference proceedings)  

  • 深層学習を用いた日本手話音節の適格性解析 Reviewed

    高藤朋史 知能数理*, 佐々木　裕, 三輪　誠, 原　大介

    言語処理学会第25回年次大会発表論文集   486 - 489   2019.3

    　More details

    Publishing type：Research paper (international conference proceedings)  

  • Deep Exhaustive Model for Nested Named Entity Recognition Reviewed

    Mohammad Golam Sohrab **, Makoto Miwa

    Proceedings of 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP 2018)   2018.11

    　More details

    Authorship：Last author   Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • What makes syllables well-formed or ill-formed in Japanese Sign Language Reviewed

    Daisuke Hara , Makoto Miwa

    The 13th High Desert Linguistics Society Conference(HDLS13)   49 - 50   2018.11

    　More details

    Authorship：Last author   Publishing type：Research paper (international conference proceedings)  

  • Investigating Domain-Specific Information for Neural Coreference Resolution on Biomedical Texts Reviewed

    Long Trieu **, Nhung Nguyen **, Makoto Miwa, Sophia Ananiadou **

    Proceedings of the 2018 Workshop on Biomedical Natural Language Processing (BioNLP 2018)   183 - 188   2018.7

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • Enhancing Drug-Drug Interaction Extraction from Texts by Molecular Structure Information Reviewed

    Masaki Asada*, Makoto Miwa, Yutaka Sasaki

    Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (ACL 2018)   680 - 685   2018.7

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • A Walk-based Model on Entity Graphs for Relation Extraction Reviewed

    Fenia Christopoulou **, Makoto Miwa, Sophia Ananiadou **

    Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (ACL 2018)   81 - 88   2018.7

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • A neural layered model for nested named entity recognition Reviewed

    Meizhi Ju **, Makoto Miwa, Sophia Ananiadou **

    Proceedings of the 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies   1446 - 1459   2018.6

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • EDGE2VEC: Edge Representations for Large-Scale Scalable Hierarchical Learning Reviewed

    Mohammad Golam Sohrab **, Toru Nakata **, Makoto Miwa, Yutaka Sasaki

    Computación y Sistemas (CyS)   21 ( 4 )   569 - 579   2017.12

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：Centro de Investigacion en Computación, IPN  

  • Analyzing well-formedness of syllables in Japanese Sign Language Reviewed

    Satoshi Yawata*, Makoto Miwa, Yutaka Sasaki, 原　大介

    Proceedings of the 8th International Joint Conference on Natural Language Processing (IJCNLP 2017)   26 - 30   2017.11

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • Utilizing visual forms of Japanese characters for neural review classification Reviewed

    Yota Toyama*, Makoto Miwa, Yutaka Sasaki

    Proceedings of the 8th International Joint Conference on Natural Language Processing (IJCNLP 2017)   378 - 382   2017.11

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • Extracting Drug-Drug Interactions with Attention CNNs Reviewed

    Masaki Asada*, Makoto Miwa, Yutaka Sasaki

    Proceedings of the BioNLP 2017 workshop   9 - 18   2017.8

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • TTI-COIN at SemEval-2017 Task 10: Investigating Embeddings for End-to-End Relation Extraction from Scientific Papers Reviewed

    Tomoki Tsujimura*, Makoto Miwa, Yutaka Sasaki

    Proceedings of the 11th International Workshop on Semantic Evaluations (SemEval-2017)   985 - 989   2017.8

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • Bib2vec: Embedding-based Search System for Bibliographic Information Reviewed

    Takuma Yoneda*, Koki Mori*, Makoto Miwa, Yutaka Sasaki

    Proceedings of the EACL 2017 Software Demonstrations   112 - 115   2017.4

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • Distributional Hypernym Generation by Jointly Learning Clusters and Projections Reviewed

    Josuke Yamane*, Tomoya Takatani **, Hitoshi Yamada **, Makoto Miwa, Yutaka Sasaki

    Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers   1871 - 1879   2016.12

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：International Committee on Computational Linguistics (ICCL)  

  • IN-DEDUCTIVE and DAG-Tree Approaches for Large-Scale Extreme Multi-label Hierarchical Text Classification Reviewed

    マハマド　ゴラム　ソフラブ, Makoto Miwa, Yutaka Sasaki

    Polibits   54   61 - 70   2016.10

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：Centro de Innovación y Desarrollo Tecnológico en Cómputo (CIDETEC)  

  • End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures Reviewed

    Makoto Miwa, Mohit Bansal TTIC*

    Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (ACL 2016)   1105 - 1116   2016.8

    　More details

    Authorship：Lead author   Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

  • Topic Detection Using Paragraph Vectors to Support Active Learning in Systematic Reviews Reviewed

    Kazuma Hashimoto **, Georgios Kontonatsios **, Makoto Miwa, Sophia Ananiadou **

    Journal of Biomedical Informatics   62   59 - 65   2016.8

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：Elsevier  

  • Ensemble Classification of Grants using LDA-based features Reviewed

    Ioannis Korkontzelos **, Beverley Thomas **, Makoto Miwa, Sophia Ananiadou **

    Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC 2016)   1288 - 1294   2016.5

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：European Language Resources Association (ELRA)  

  • Text Mining for Semantic Search in Europe PubMed Central Labs Reviewed International coauthorship

    W. J. Black, A. Rowley, J. McNaught, S. Ananiadou, M. Miwa

    Working with Text: Tools, Techniques and Approaches for Text Mining   111 - 131   2016.1 （ ISBN:9781843347491 ）

    　More details

    Language：English   Publishing type：Part of collection (book)  

    This chapter describes an implemented and publicly available search aid designed for use in the context of a standard full-text retrieval service, which automatically suggests questions on the basis of what has been entered in the search engine’s query field. The service is developed on the basis of a content analysis achieved by merging information extraction of biomedical named entities with a syntactic analysis of the full text of an entire collection of scientific papers. We discuss the design and implementation of the system in contrast with alternative ways of providing a search application based on text mining for events of significance in biomedical sciences, and evaluate characteristics of the system against criteria established in collaboration with sample users.

    DOI： 10.1016/B978-1-84334-749-1.00005-6

  • Stacking Approach to Temporal Relation Classification with Temporal Inference Reviewed

    Natsuda Laokulrat The University of Tokyo*, Makoto Miwa, Yoshimasa Tsuruoka The University of Tokyo*

    自然言語処理   22 ( 3 )   171 - 196   2015.9 （   ISSN:1340-7619 ）

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：言語処理学会  

  • Identifying synonymy between relational phrases using word embeddings Reviewed

    Nhung T. H. Nguyen JAIST*, Makoto Miwa, Yoshimasa Tsuruoka The University of Tokyo*, Satoshi Tojo JAIST*

    Journal of Biomedical Informatics   56   94 - 102   2015.8

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：ELSEVIER  

  • Task-Oriented Learning of Word Embeddings for Semantic Relation Classification Reviewed

    Kazuma Hashimoto The University of Tokyo*, Pontus Stenetorp University College London*, Makoto Miwa, Yoshimasa Tsuruoka The University of Tokyo*

    Proceedings of the 19th Conference on Computational Language Learning   268 - 278   2015.7

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：ACL  

  • Adaptable, High Recall, Event Extraction System with Minimal Configuration Reviewed

    Makoto Miwa, Sophia Ananiadou University of Manchester*

    BMC Bioinformatics   16 ( Suppl 10 )   S7   2015.6

    　More details

    Authorship：Lead author   Publishing type：Research paper (scientific journal)   Publisher：BMC  

  • Word Embedding-based Antonym Detection using Thesauri and Distributional Information Reviewed

    Masataka Ono 知能数理*, Makoto Miwa, Yutaka Sasaki

    Proceedings of the 2015 Annual Conference of the North American Chapter of the Association for Computational Linguistics - Human Language Technologies (NAACL-HLT 2015)   984 - 989   2015.5

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：ACL  

  • Centroid-Means-embedding: An Approach to Infusing Word Embeddings into Features for Text Classifcation Reviewed

    マハマド　ゴラム　ソフラブ, Makoto Miwa, Yutaka Sasaki

    Advances in Knowledge Discovery and Data Mining   LNAI ( 9077 )   289 - 300   2015.4

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：Pacific-Asia Knowledge Discovery and Data Mining  

  • Wide-Coverage Relation Extraction from MEDLINE Using Deep Syntax Reviewed

    Nhung T. H. Nguyen JAIST*, Makoto Miwa, Yoshimasa Tsuruoka The University of Tokyo*, Takashi Chikayama The University of Tokyo*, Satoshi Tojo JAIST*

    BMC Bioinformatics   16 ( 1 )   107   2015.4 （   ISSN:1471-2105 ）

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：BMC  

  • Using text mining for study identification in systematic reviews: a systematic review of current approaches Reviewed

    Alison O'Mara-Eves Univ. of London*, James Thomas Univ. of London*, John McNaught Univ. of Manchester*, Makoto Miwa, Sophia Ananiadou Univ. of Manchester*

    Systematic Reviews   4 ( 1 )   5   2015.1

    　More details

    Publishing type：Research paper (scientific journal)  

  • Centroid-Means-Embedding: An Approach to Infusing Word Embeddings into Features for Text Classification Reviewed

    Yutaka Sasaki, Makoto Miwa, マハマド　ゴラム　ソフラブ

    The 19th Pacific-Asia Conference on Knowledge Discovery and Data Mining   2015

    　More details

    Publishing type：Research paper (international conference proceedings)  

  • 多人数性を分割した教師付き学習による4人麻雀プログラムの実現 Reviewed

    水上 直紀 (東大)*, 中張 遼太郎 (東大)*, 浦 晃 (東大)*, 三輪　誠, 鶴岡 慶雅 (東大)*, 近山 隆 (東大)*

    情報処理学会論文誌   55 ( 11 )   2410 - 2420   2014.11

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：情報処理学会  

  • 対数線形言語モデルを用いた将棋解説文の自動生成 Reviewed

    亀甲 博貴 (東大)*, 三輪　誠, 鶴岡 慶雅 (東大)*, 森 信介 (京大)*, 近山 隆 (東大)*

    情報処理学会論文誌   55 ( 11 )   2431 - 2440   2014.11 （   ISSN:1882-7764 ）

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：情報処理学会  

  • Modeling Joint Entity and Relation Extraction with Table Representation Reviewed

    Makoto Miwa, Yutaka Sasaki

    Proceedings of 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP 2014)   2014.10

    　More details

    Authorship：Lead author   Publishing type：Research paper (international conference proceedings)  

  • Reducing systematic review workload through certainty-based screening Reviewed

    Makoto Miwa, James Thomas Univ. of London*, Alison O'Mara-Eves Univ. of London*, Sophia Ananiadou Univ. of Manchester*

    Journal of Biomedical Informatics   51   242 - 253   2014.10

    　More details

    Authorship：Lead author   Publishing type：Research paper (scientific journal)   Publisher：Academic Press  

  • Comparable Study of Event Extraction in Newswire and Biomedical Domains Reviewed

    Makoto Miwa, Paul Thompson (Univ. of Manchester)*, Ioannis Korkontzelos (Univ. of Manchester)*, Sophia Ananiadou (Univ. of Manchester)*

    Proceedings of the 25th International Conference on Computational Linguistics (COLING 2014)   2270 - 2279   2014.8

    　More details

    Authorship：Lead author   Publishing type：Research paper (international conference proceedings)  

▼display all

MISC

  • Extracting Synthetic Material Names and Characteristic Values from Inorganic Material Science Papers using a Named Entity Tagger

    KUNIYOSHI Fusataka, OSAWA Jun, MIWA Makoto

    Proceedings of the Annual Conference of JSAI   JSAI2021   4F4GS10o03 - 4F4GS10o03   2021

    　More details

    Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：The Japanese Society for Artificial Intelligence  

    In the field of inorganic materials, efforts are being made to discover materials with high properties in a short time by referring to statistical data that links synthetic material names with their property values. However, there are few large-scale databases that link synthetic material names with their property values. In this study, we focus on the extraction of information from academic papers. We first improve the existing annotation scheme for extracting the synthetic material names of batteries described in natural language in papers by adding a new property value label to extract material names and their property values simultaneously. By using our annotation scheme, we built a corpus which includes 836 paragraphs extracted from 301 papers to train a named entity extraction model. The evaluation results show that our named entity extraction model has high extraction performance. In addition, we extracted pairs of synthetic material names and property values from 24,415 material articles using the named entity extraction model. Finally, the extraction results are visualized in a simple and the trend of the materials in each period is discussed, demonstrating the usefulness of a large-scale database consisting of the pairs of synthetic processes and their property values.

    DOI： 10.11517/pjsai.jsai2021.0_4f4gs10o03

  • Annotation and Classification of Graphs of Battery Property in Material Science Literature

    IINUMA Naoki, KUNIYOSHI Fusataka, OZAWA Jun, MIWA Makoto

    Proceedings of the Annual Conference of JSAI   JSAI2021   2Xin515 - 2Xin515   2021

    　More details

    Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：The Japanese Society for Artificial Intelligence  

    The task of extracting material property values described in the text of material papers has attracted much attention among materials researchers. However, many property values cannot be extracted by natural language processing alone because property values are often described in graphs rather than in the text in materials papers. In this study, to extract property values from graphs, we constructed a dataset by classifying graphs of property values into classes based on various property conditions such as temperature and time. The dataset was constructed by extracting graph images from a large collection of journal data in the field of materials and utilizing crowdsourcing to annotate the images in a short period of time. In addition, we built several deep learning models and trained and evaluated them on the dataset. As a result, we confirmed the usefulness of our dataset for classifying graphs of property values using deep learning models.

    DOI： 10.11517/pjsai.jsai2021.0_2xin515

  • Target material extraction from literature describing synthesis procedures in the field of inorganic materials science

    MAKINO Kohei, KUNIYOSHI Fusataka, OZAWA Jun, MIWA Makoto

    Proceedings of the Annual Conference of JSAI   JSAI2021   2Xin516 - 2Xin516   2021

    　More details

    Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：The Japanese Society for Artificial Intelligence  

    In the inorganic materials field, research has been conducted to extract target materials, which are the synthetic materials claimed in the papers, to focus on synthetic materials and analyze their physical properties. For the extraction, there is a question whether the conventional named entity recognition systems can extract such target materials. In this study, we built a corpus of papers labeled only with the target materials and applied a deep learning modes, which have shown high performance in conventional named entity extraction extraction tasks, to the corpus to evaluate the extraction performance of the target materials. As a result, we found that the performance of the deep learning model in extracting target materials was lower than that reported in other named entity recognition tasks. We attribute this to the fact that the conventional named entity recognition task settings are not suitable for the task of extracting target materials from articles, and we discuss the shortcomings of the existing task settings and ways to improve them.

    DOI： 10.11517/pjsai.jsai2021.0_2xin516

  • Poincare GloVeベクトルのレトロフィッティング

    村瀬敦也, 三輪誠, 佐々木裕

    言語処理学会年次大会発表論文集(Web)   26th   2020 （   ISSN:2188-4420 ）

    　More details

    Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

  • Synthesis Process Paragraph Extraction from Scientific Literature of Inorganic Material

    MAKINO Kohei, KUNIYOSHI Fusataka, OZAWA Jun, MIWA Makoto

    Proceedings of the Annual Conference of JSAI   JSAI2020   4Rin112 - 4Rin112   2020

    　More details

    Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)   Publisher：The Japanese Society for Artificial Intelligence  

    In the field of inorganic materials, there is a need for a system that supports development by analyzing synthesis processes described in a large number of papers. In order to realize the system, it is necessary to extract the part where synthesis processes are described from the papers. We propose a tool that extracts paragraphs describing synthesis processes from papers in the PDF format. We develop the tool by combining a deep learning-based sentence classifier that determines whether each sentence includes synthesis processes or not and a paragraph detector using the sentence classifier. In the experiment, we evaluated our tool on manually-labeled 300 papers. As a result, our tool performed well in both classifying sentences and detecting paragraphs. This result shows that the proposed tool is useful in extracting paragraphs on synthesis processes.

    DOI： 10.11517/pjsai.jsai2020.0_4rin112

  • 人体姿勢アノテーション困難な映像における類似姿勢学習の有用性

    村上達哉, 三輪誠, 浮田宗伯

    情報処理学会研究報告(Web)   2018 ( CVIM-212 )   Vol.2018‐CVIM‐212,No.13,1‐7 (WEB ONLY)   2018.5

    　More details

    Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

  • n‐gram素性に対する注意機構を利用したニューラルネットによる単語穴埋め

    森洸樹, 三輪誠, 佐々木裕

    言語処理学会年次大会発表論文集(Web)   23rd   ROMBUNNO.P8‐3 (WEB ONLY)   2017 （   ISSN:2188-4420 ）

    　More details

    Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

  • 文書の柔軟な検索に向けたキーワードと文書の意味表現の獲得

    三輪誠, 三輪誠

    豊田研究報告   ( 69 )   149‐150   2016.5 （   ISSN:0372-039X ）

    　More details

    Authorship：Lead author,　Last author,　Corresponding author   Language：Japanese   Publishing type：Internal/External technical report, pre-print, etc.  

▼display all

Books

  • Working with Text: Tools, Techniques and Approaches for Text Mining

    William J. Black *, Andrew Rowley *, Makoto Miwa, John McNaught *, Sophia Ananiadou *（Text Mining for Semantic Search in Europe PubMed Central Labs）

    Chandos Publishing  2016.7  （ ISBN:978-1-8433-4749-1 ）

    　More details

    Responsible for pages：111--132   Book type：Scholarly book

Presentations

  • IDレベル関係抽出における不要な文の自動選択

    辻村有輝*, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 訓練可能なk近傍Retrieverで関係抽出事例を導入したニューラルプロンプティング

    牧野晃平*, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 他文書の予測を知識グラフに蓄積・利用する文書単位関係抽出

    松原拓磨*, 辻村有輝, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • CVAEによる複数データセットからの固有表現抽出

    大井拓*, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 文献グラフにおける多項関係の埋め込み

    井田龍希*, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 複数の形式・表現の質問を利用した多角的な関係抽出

    山田晃士*, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • CVAEによる複数データセットからの固有表現抽出

    大井拓*, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 表層が同じ文字列の同一性を表現した深層固有表現抽出

    吉村貴紀*, 牧野晃平, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 超伝導材料の転移温度予測における事例間の繋がりを考慮した知識グラフの有効性の調査

    吉野草太*, 旭　良司, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 固有表現抽出における大規模言語モデルのLoRAファインチューニングの学習設定の調査

    鬼頭泰清*, 牧野晃平, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • イベントの発生条件のアノテーションと条件の予測性能評価

    市村裕章*, 三輪　誠, 佐々木　裕

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 外国手話データセットを活用した日本手話動画からの音節構成要素認識

    木全純大*, 三輪　誠, 佐々木　裕, 原　大介

    第30回言語処理学会年次大会  ( 神戸 )   2024.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • コーパス内の関係間の相互作用を考慮した関係抽出

    2023.8 

    　More details

    Presentation type：Oral presentation (general)  

  • Biomedical Document Classification with Literature Graph Representations of Bibliographies and Entities International conference

    BioNLP 2023  ( Online )   2023.7  Association for Computational Linguistics

    　More details

    Presentation type：Oral presentation (general)  

    This paper proposes a new document classification method that incorporates the representations of a literature graph created from bibliographic and entity information. Recently, document classification performance has been significantly improved with large pre-trained language models; however, there still remain documents that are difficult to classify. External information, such as bibliographic information, citation links, descriptions of entities, and medical taxonomies, has been considered one of the keys to dealing with such documents in document classification. Although several document classification methods using external information have been proposed, they only consider limited relationships, e.g., word co-occurrence and citation relationships. However, there are multiple types of external information. To overcome the limitation of the conventional use of external information, we propose a document classification model that simultaneously considers bibliographic and entity information to deeply model the relationships among documents using the representations of the literature graph. The experimental results show that our proposed method outperforms existing methods on two document classification datasets in the biomedical domain with the help of the literature graph.

  • Distantly supervised document-level biomedical relation extraction with neighborhood knowledge graphs International conference

    BioNLP 2023  ( Online )   2023.7  Association for Computational Linguistics

    　More details

    Presentation type：Oral presentation (general)  

    We propose a novel distantly supervised document-level biomedical relation extraction model that uses partial knowledge graphs that include the graph neighborhood of the entities appearing in each input document. Most conventional distantly supervised relation extraction methods use only the entity relations automatically annotated by using knowledge base entries. They do not fully utilize the rich information in the knowledge base, such as entities other than the target entities and the network of heterogeneous entities defined in the knowledge base. To address this issue, our model integrates the representations of the entities acquired from the neighborhood knowledge graphs with the representations of the input document. We conducted experiments on the ChemDisGene dataset using Comparative Toxicogenomics Database (CTD) for document-level relation extraction with respect to interactions between drugs, diseases, and genes. Experimental results confirmed the performance improvement by integrating entities and their neighborhood biochemical information from the knowledge base.

  • Biomedical Relation Extraction with Entity Type Markers and Relation-specific Question Answering International conference

    BioNLP 2023  ( Online )   2023.7  Association for Computational Linguistics

    　More details

    Presentation type：Oral presentation (general)  

    Recently, several methods have tackled the relation extraction task with QA and have shown successful results. However, the effectiveness of existing methods in specific domains, such as the biomedical domain, is yet to be verified. When there are multiple entity pairs that share an entity in a sentence, a QA-based relation extraction model that outputs only one single answer to a given question may not extract desired relations. In addition, these methods employ QA models that are not tuned for relation extraction. To address these issues, we first extend and apply a span QA-based relation extraction method to the drug-protein relation extraction by creating question templates and incorporating entity type markers. We further propose a binary QA-based method that directly uses the entity information available in the relation extraction task. The experimental results on the DrugProt dataset show that our QA-based methods, especially the proposed binary QA method, are effective for drug-protein relation extraction.

  • 時系列予測における人工データを用いたデータ拡張

    2023.6 

    　More details

    Presentation type：Oral presentation (general)  

  • 複数の質問形式を利用した分類型の質問応答による薬物タンパク質間関係抽出

    山田晃士*, 三輪　誠, 佐々木　裕

    第29回言語処理学会年次大会  ( 沖縄 )   2023.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 文書外の書誌情報と用語情報を組み込んだ文書分類

    井田龍希*, 三輪　誠, 佐々木　裕

    第29回言語処理学会年次大会  ( 沖縄 )   2023.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • ラベル内容のエンコードとラベル間の制約に基づく補助コーパスを用いた固有表現抽出

    大井拓*, 三輪　誠, 佐々木　裕

    第29回言語処理学会年次大会  ( 沖縄 )   2023.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 近傍知識グラフからの埋め込みを統合利用する文書からの遠距離教師あり関係抽出

    松原拓磨*, 三輪　誠, 佐々木　裕

    第29回言語処理学会年次大会  ( 沖縄 )   2023.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • テキスト情報の表現を利用した文献グラフの表現学習

    片桐脩那*, 井田龍希*, 三輪　誠, 佐々木　裕

    第29回言語処理学会年次大会  ( 沖縄 )   2023.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 日本手話の音声的手型と音素的手型

    原　大介, 三輪　誠

    日本手話学会第48回大会  ( 東京大学・先端科学技術センター )   2022.12  日本手話学会

    　More details

    Presentation type：Oral presentation (general)  

  • Synthetic Data Augmentation for Time Series Forecasting International conference

    Women in Machine Learning (WiML) Workshop 2022  ( Online )   2022.12  Women in Machine Learning

    　More details

    Presentation type：Oral presentation (general)  

  • 知識ベースを用いた科学文献からの情報抽出 Invited

    三輪　誠

    2022年度 データサイエンス特別講義  ( 奈良先端科学技術大学院大学 )   2022.11  奈良先端科学技術大学院大学

    　More details

    Presentation type：Oral presentation (invited, special)  

  • Span-based and Question Answering-based Medication Event Extraction International conference

    2022 n2c2 Shared Task and Workshop  ( Washington D.C. )   2022.11  n2c2

    　More details

    Presentation type：Oral presentation (general)  

  • Information Extraction from Text with Heterogeneous Knowledgebase Information Invited International conference

    Sixth International Workshop on Symbolic-Neural Learning (SNL2022)  ( Toyota Technological Institute, Nagoya )   2022.7  SNL2022 Organizing Committee

    　More details

    Presentation type：Oral presentation (invited, special)  

  • Improving Supervised Drug-Protein Relation Extraction with Distantly Supervised Models International conference

    BioNLP 2022  ( Online )   2022.5  Association for Computational Linguistics

    　More details

    Presentation type：Oral presentation (general)  

  • 遠距離教師データの特徴表現を活用した薬物タンパク質間関係抽出

    飯沼直己*, 三輪　誠, 佐々木　裕

    言語処理学会第28回年次大会  ( オンライン )   2022.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 項の表現に着目した質問応答による関係分類

    山田晃士*, 三輪　誠, 佐々木　裕

    言語処理学会第28回年次大会  ( オンライン )   2022.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 単語を共有する文書グラフを用いた文書分類

    井田龍希*, 三輪　誠, 佐々木　裕

    言語処理学会第28回年次大会  ( オンライン )   2022.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 関係間の関係性を考慮した時間関係グラフ改善のためのグローバル反復辺編集器

    牧野晃平*, 三輪　誠, 佐々木　裕

    言語処理学会第28回年次大会  ( オンライン )   2022.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • タスク指向対話システムの外部表知識の参照能力向上

    深谷竜暉*, 三輪　誠, 佐々木　裕

    言語処理学会第28回年次大会  ( オンライン )   2022.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 交通に関する知識グラフを用いた運転免許試験問題の解法

    相川渉*, 三輪　誠, 佐々木　裕

    言語処理学会第28回年次大会  ( オンライン )   2022.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 生化学分野のフルペーパーを対象としたリンキングと索引付け

    辻村有輝*, 井田龍希*, 三輪　誠, 佐々木　裕

    言語処理学会第28回年次大会  ( オンライン )   2022.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 薬学知識グラフ上のヘテロな情報を利用した文献からの薬物相互作用抽出

    浅田真生*, 三輪　誠, 佐々木　裕

    言語処理学会第28回年次大会  ( オンライン )   2022.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 日本手話における手型変化

    原　大介, 三輪　誠

    日本手話学会第47回大会  ( 大阪府立男女共同参画・青少年センター )   2021.12  日本手話学会

    　More details

    Presentation type：Oral presentation (general)  

  • Information Extraction from Texts Using Heterogeneous Information Invited International conference

    BioNLP 2021  ( Virtual )   2021.6  Association for Computational Linguistics

    　More details

    Presentation type：Oral presentation (invited, special)  

  • 外部知識を利用したテキストからの情報抽出 Invited

    三輪　誠

    京都大学学術情報メディアセンターセミナー  ( オンライン )   2021.5  京都大学学術情報メディアセンター

    　More details

    Presentation type：Oral presentation (invited, special)  

  • 組み込み・除外判定を機械読解により実現した系統的レビュー

    佐々木　裕, 三輪　誠, 安部 賀央里 名市大**, 頭金 正博 名市大**

    第248回自然言語処理・第226回コンピュータビジョンとイメージメディア合同研究発表会  ( オンライン )   2021.5  情報処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 辺編集による文書レベルの関係グラフ構築

    牧野晃平 知能数理*, 三輪　誠, 佐々木　裕

    言語処理学会第27回年次大会  ( オンライン )   2021.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 機械加工文書における用語入れ子構造とトリガワードを考慮した用語関係同時抽出

    稲熊陸 知能数理*, 小島大*, 東孝幸*, 三輪　誠, 古谷　克司, 佐々木　裕

    言語処理学会第27回年次大会  ( オンライン )   2021.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • Extrametricality of the initial location in the type-III syllable of Japanese Sign Language International conference

    28th Japanese/Korean Linguistics Virtual Conference (JK28)  ( Online )   2020.9  Japanese/Korean Linguistics Conference

    　More details

    Presentation type：Oral presentation (general)  

  • 小型UGV自動走行のための深層強化学習手法の比較

    藤井匠透 知能数理*, 三輪　誠, 佐々木　裕

    第34回人工知能学会全国大会  ( オンライン )   2020.6  人工知能学会

    　More details

    Presentation type：Oral presentation (general)  

  • 二段階学習と概念クラスを用いた医療固有表現の正規化

    茂里憲之*, 辻村有輝*, 三輪　誠, 佐々木　裕

    言語処理学会第26回年次大会  ( オンライン )   2020.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 入れ子構造を考慮した機械加工用語抽出

    稲熊陸*, 小島大 **, 東孝幸 **, 三輪　誠, 古谷　克司, 佐々木　裕

    言語処理学会第26回年次大会  ( オンライン )   2020.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • オントロジー形式アノテーションを対象とした交通用語・関係抽出と正誤問題の回答

    鈴木直樹*, サーヴォン　ブー, 三輪　誠, 佐々木　裕

    言語処理学会第26回年次大会  ( オンライン )   2020.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 無機材料文献からの合成プロセス抽出のための関係抽出

    牧野晃平*, 國吉房貴*, 小澤順 **, 三輪　誠

    言語処理学会第26回年次大会  ( オンライン )   2020.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • コーディングと動画を併用した日本手話音節の適格性予測

    高藤朋史*, 三輪　誠, 佐々木　裕, 原　大介

    言語処理学会第26回年次大会  ( オンライン )   2020.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • オントロジー形式による交通関係アノテーション

    サーヴォン　ブー, 鈴木直樹*, 三輪　誠, 佐々木　裕

    言語処理学会第26回年次大会  ( オンライン )   2020.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • Poincaré GloVe ベクトルのレトロフィッティング

    村瀬敦也*, 三輪　誠, 佐々木　裕

    言語処理学会第26回年次大会  ( オンライン )   2020.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 遠距離教師データを援用した教師あり薬物タンパク質間相互作用抽出

    飯沼直己*, 三輪　誠, 佐々木　裕

    言語処理学会第26回年次大会  ( オンライン )   2020.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • TTI-COIN at n2c2 2019 Track 3: Neural Medical Concept Normalization with Two-Step Training International conference

    2019 n2c2/OHNLP Workshop  ( Online )   2019.11 

    　More details

    Presentation type：Oral presentation (general)  

  • 深層学習を用いた車両運動性能の代理モデルの開発

    牧野晃平 知能数理*, 三輪　誠, 新谷浩平 トヨタ自動車*, 阿部充浩 トヨタ自動車*, 佐々木　裕

    第29回設計工学・システム部門講演会  ( 東北大学 )   2019.9  日本機械学会

    　More details

    Presentation type：Oral presentation (general)  

  • The phonotactics of type-III syllables of Japanese Sign Language International conference

    Theoretical Issues in Sign Language Research 13(TISLR13)  ( The University of Hamburg, Germany )   2019.9  Sign Language Linguistics Society

    　More details

    Presentation type：Oral presentation (general)  

  • 係り受け木上の不要な情報にマスクを行う関係分類

    辻村有輝*, 三輪　誠, 佐々木　裕

    NLP若手の会 第14回シンポジウム  ( ホテルエミシア札幌 )   2019.8  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 薬物データベースを統合的に利用する薬物相互作用抽出

    浅田真生*, 三輪　誠, 佐々木　裕

    NLP若手の会 第14回シンポジウム  ( ホテルエミシア札幌 )   2019.8  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • Sequence Generation by Sequential Time-Point GAN International conference

    Third International Workshop on Symbolic-Neural Learning  ( Tokyo )   2019.7  Organizing Committee for Symbolic-Neural Learning

    　More details

    Presentation type：Oral presentation (general)  

  • Using External DB Knowledge in Neural DDI Extraction International conference

    Third International Workshop on Symbolic-Neural Learning  ( Tokyo )   2019.7  Organizing Committee for Symbolic-Neural Learning

    　More details

    Presentation type：Oral presentation (general)  

  • Masking Unnecessary Information in Dependency Trees for Neural Relation Classification International conference

    Tomoki Tsujimura*, Makoto Miwa, Yutaka Sasaki

    Third International Workshop on Symboic-Neural Learning  2019.7 

    　More details

    Presentation type：Oral presentation (general)  

    In relation classification, the mention about the relation often exists in the shortest dependency path between target entities and omitting tokens outside the shortest path from the input of the relation classification model improves generalization ability. However, it is a heuristic rule and inflexible to unexpected relations such as relations that require information outside of the path and relations not directly mentioned. We propose a novel masking mechanism for neural relation classification that learns to mask unnecessary nodes in dependency trees in an end-to-end manner. Our masking mechanism works as a hidden layer to drop unnecessary hidden vectors at the token level by discrete masks during both training and test time. Following layers process only the remaining unmasked tokens and aggregate them with an attention mechanism to represent relations. We show that the relation classification model with our method performs the results comparable to the one obtained from the model using the shortest path heuristic. We also investigate the differences in the remaining tokens between the shortest path and the learned masks.

  • The well-formedness and the ill-formedness of the JSL type-Ⅲ syllables International conference

    Daisuke Hara, Makoto Miwa

    The Chicago Linguistic Society 55th Annual Meeting  2019.5 

    　More details

    Presentation type：Oral presentation (general)  

  • データベースの説明文を利用した薬物相互作用抽出

    浅田真生*, 三輪　誠, 佐々木　裕

    言語処理学会第25回年次大会  ( 名古屋大学 )   2019.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 全文字n-gramを考慮したDilated CNNを用いた単語分割

    山口修平*, 三輪　誠, 佐々木　裕

    言語処理学会第25回年次大会  ( 名古屋大学 )   2019.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • CNNを用いた交通教則からの交通用語間関係抽出

    八木智也*, 三輪　誠, 佐々木　裕

    言語処理学会第25回年次大会  ( 名古屋大学 )   2019.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 深層学習を用いた日本手話音節の適格性解析

    高藤朋史 知能数理*, 三輪　誠, 佐々木　裕, 原　大介

    言語処理学会第25回年次大会  ( 名古屋大学 )   2019.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 遠距離教師データを援用した教師有り薬物タンパク質間相互作用抽出

    矢島雄樹*, 三輪　誠, 佐々木　裕

    言語処理学会第25回年次大会  ( 名古屋大学 )   2019.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 係り受け木上の不要な情報にマスクを行う関係分類

    辻村有輝*, 三輪　誠, 佐々木　裕

    言語処理学会第25回年次大会  ( 名古屋大学 )   2019.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 文献からの情報抽出の現在と今後 Invited

    三輪　誠

    富士通研究所講演会  ( 富士通研究所 )   2018.12  富士通研究所

    　More details

    Presentation type：Oral presentation (invited, special)  

  • What makes syllables well-formed or ill-formed in Japanese Sign Language International conference

    Daisuke Hara, Makoto Miwa

    The 13th High Desert Linguistics Society Conference  2018.11 

    　More details

    Presentation type：Oral presentation (general)  

    The purpose of this study is to try to find the rules that function to distinguish the well-formed and ill-formed syllables of Japanese Sign Language (JSL). In other words, gto establish the phonotactics of JSL. To find out what makes syllables well-formed or ill-formed, we have adopted two methodological approaches to this problem: one is a conventional linguistic, i.e., descriptive, observation, and the other is a kind of new approach of using machine learning algorithms, in which the authors make a computer learn differences between the well-formed and the ill-formed JSL syllables and find features effective to distinguish them. This poster presentation is a progress report of this two-sided approach, especially the one from the linguistic side.

  • ニューラル機械翻訳に適応する句構造構築モデル

    野中舜介*, 三輪　誠, 佐々木　裕

    NLP若手の会 第13回シンポジウム  ( 喜代美山荘　花樹海 )   2018.8  NLP 若手の会第13回シンポジウム運営委員

    　More details

    Presentation type：Oral presentation (general)  

  • Prediction Forging Dies via Generative Adversarial Networks for Pairs in Sequences International conference

    Hayato Futase*, Makoto Miwa, Yutaka Sasaki

    2nd International Workshop on Symbolic-Neural Learning  2018.7 

    　More details

    Presentation type：Oral presentation (general)  

    Preliminary report on devising a new GAN called PairGan for predicting forging dies.

  • Neural Methods for Semantic Relation Extraction from Texts and Databases Invited International conference

    Makoto Miwa

    Second International Workshop on Symbolic-Neural Learning (SNL-2018)  2018.7 

    　More details

    Presentation type：Oral presentation (invited, special)  

  • Semantic Graph Embeddings and a Neural Language Model for Word Sense Disambiguation International conference

    マーク エブラル, Makoto Miwa, Yutaka Sasaki

    2nd International Workshop on Symbolic-Neural Learning  2018.7 

    　More details

    Presentation type：Oral presentation (general)  

    Applying Poincare embeddings to Word Sense Disambiguation

  • 深層情報抽出に向けて Invited

    三輪　誠

    第2回 AAMT／Japio特許翻訳研究会  ( キャンパス・イノベーションセンター東京 )   2018.6  AAMT／Japio特許翻訳研究会

    　More details

    Presentation type：Oral presentation (invited, special)  

  • コーパスを通してみる日本手話音節の（不）適格性

    原　大介, 三輪　誠

    電子情報通信学会・リアルタイムコミュニケーション言語研究会（LARC)  ( 鹿児島 )   2018.6 

    　More details

    Presentation type：Oral presentation (general)  

  • 文献情報の多様な要素を考慮したベクトル表現獲得

    米田拓真*, 三輪　誠, 佐々木　裕

    第24回言語処理学会年次大会  ( 岡山コンベンションセンター（ママカリフォーラム），岡山市 )   2018.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 分子構造を用いた文書からの薬物相互作用抽出

    浅田真生 **, 三輪　誠, 佐々木　裕

    第24回言語処理学会年次大会  ( 岡山コンベンションセンター（ママカリフォーラム），岡山市 )   2018.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 薬物の系統的レビューにおける選択基準ベクトルの利用

    佐々木　裕, 三輪　誠, 安部賀央里 **, 頭金正博 **

    第24回言語処理学会年次大会  ( 岡山コンベンションセンター（ママカリフォーラム），岡山市 )   2018.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 分割候補の探索をともなうニューラル日本語単語分割

    山口修平 **, 三輪　誠, 佐々木　裕

    第24回言語処理学会年次大会  ( 岡山コンベンションセンター（ママカリフォーラム），岡山市 )   2018.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 双方向性と非線形性を考慮した上位語・下位語関係の推定

    山根丈亮 **, 高谷智哉 **, 山田整 **, 三輪　誠, 佐々木　裕

    第24回言語処理学会年次大会  ( 岡山コンベンションセンター（ママカリフォーラム），岡山市 )   2018.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • ニューラル文生成モデルを利用した英文読解問題の自動解法

    加藤秀大*, 三輪　誠, 佐々木　裕

    第24回言語処理学会年次大会  ( 岡山コンベンションセンター（ママカリフォーラム），岡山市 )   2018.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • ニューラルVQAのTOEIC写真問題への領域適応

    高里盛良*, 三輪　誠, 佐々木　裕

    第24回言語処理学会年次大会  ( 岡山コンベンションセンター（ママカリフォーラム），岡山市 )   2018.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 表現学習：柔軟な知識をもったコンピュータに向けて Invited

    三輪　誠

    第13回ジョイントCS セミナー  ( 豊田工業大学 )   2017.11  豊田工業大学・豊田工業大学シカゴ校

    　More details

    Presentation type：Oral presentation (invited, special)  

  • Analyzing Well-Formedness of Syllables in Japanese Sign Language International conference

    The 8th International Joint Conference on Natural Language Processing  ( Taipei, Taiwan. )   2017.11  National Taiwan Normal University and The Association for Computational Linguistics and Chinese Language Processing (ACLCLP)

    　More details

    Presentation type：Oral presentation (general)  

  • 辞書情報を用いた構造学習によるニューラル単語分割

    山口修平*, 三輪　誠, 佐々木　裕

    NLP若手の会 第12回シンポジウム   ( 沖縄かりゆしアーバンリゾート・ナハ )   2017.9  NLP若手の会

    　More details

    Presentation type：Oral presentation (general)  

  • TTI's Approaches to Symbolic-Neural Learning International conference

    First International Workshop on Symbolic-Neural Learning (SNL-2017)  ( Nagoya Congress Center (Nagoya, Japan) )   2017.7  TTI & TTIC

    　More details

    Presentation type：Oral presentation (general)  

  • The well-formedness condition of the Japanese Sign Language syllable International conference

    Language as a Form of Action, June 21- 23, 2017,   ( Rome, Italy )   2017.6  Deictic communication

    　More details

    Presentation type：Oral presentation (general)  

  • LSTMを用いた句のベクトル表現学習

    水口凱*, 高谷智哉 **, 山田整 **, 三輪　誠, 佐々木　裕

    人工知能学会全国大会  ( ウインクあいち（愛知県産業労働センター），名古屋 )   2017.5  人工知能学会

    　More details

    Presentation type：Oral presentation (general)  

  • アテンションCNNによる薬物間相互作用抽出

    浅田真生*, 三輪　誠, 佐々木　裕

    人工知能学会全国大会  ( ウインクあいち（愛知県産業労働センター），名古屋 )   2017.5  人工知能学会

    　More details

    Presentation type：Oral presentation (general)  

  • Markov Logic Network による機械加工因子間の確率推定

    佐々木　裕, 三輪　誠, 古谷　克司, 原田博正 **, 寺本一成 **

    人工知能学会全国大会  ( ウインクあいち（愛知県産業労働センター），名古屋 )   2017.5  人工知能学会

    　More details

    Presentation type：Oral presentation (general)  

  • 情報構造抽出に向けた深層学習・表現学習 Invited

    三輪　誠

    みちのく伝達学セミナー  ( 東北大学 )   2017.5  東北大学 乾・岡崎研究室

    　More details

    Presentation type：Oral presentation (invited, special)  

  • 線形化された構文情報を用いた生成型ニューラル文要約

    瀧川雅也*, 三輪　誠, 佐々木　裕

    第23回言語処理学会年次大会  ( 筑波大学 )   2017.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • ラティス構造を学習するニューラル単語分割

    山口修平 知能数理*, 山根丈亮 知能数理*, 三輪　誠, 佐々木　裕

    第23回言語処理学会年次大会  ( 筑波大学 )   2017.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 交通オントロジーに基づく質問応答データセットの構築

    高山隼矢 知能数理*, 三輪　誠, 佐々木　裕

    第23回言語処理学会年次大会  ( 筑波大学 )   2017.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • n-gram素性に対する注意機構を利用したニューラルネットによる単語穴埋め

    森洸樹 知能数理*, 三輪　誠, 佐々木　裕

    第23回言語処理学会年次大会  ( 筑波大学 )   2017.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • Learning for Information Extraction in Biomedical and General Domains Invited International conference

    BioTxtM 2016  ( Osaka International Convention Center )   2016.12  International Committee on Computational Linguistics (ICCL)

    　More details

    Presentation type：Oral presentation (invited, special)  

  • 関係抽出における係り受け木内部構造のアテンションによる重要度付け

    辻村有輝 知能数理*, 三輪　誠, 佐々木　裕

    NLP若手の会 第11回シンポジウム  ( 和歌山県西牟婁郡白浜町　ホテルシーモア )   2016.8  NLP若手の会

    　More details

    Presentation type：Oral presentation (general)  

  • レーティング予測によるフォントを基盤としたレビュー解析

    外山洋太 知能数理*, 三輪　誠, 佐々木　裕

    NLP若手の会 第11回シンポジウム  ( 和歌山県西牟婁郡白浜町　ホテルシーモア )   2016.8  NLP若手の会

    　More details

    Presentation type：Oral presentation (general)  

  • 上位語・下位語の射影関係とそのクラスタの同時学習

    山根丈亮 知能数理*, 高谷智哉 トヨタ自動車*, 山田整 トヨタ自動車*, 三輪　誠, 佐々木　裕

    第22回言語処理学会年次大会  ( 東北大学 )   2016.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • Trigger Wordと部分文字列を用いた機械加工用語の関係抽出

    増田和浩 知能数理*, 寺本一成 豊田中研*, 三輪　誠, 古谷　克司, 佐々木　裕

    第22回言語処理学会年次大会  ( 東北大学 )   2016.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • Word Embeddings in Large Scale Deep Architechture Learning

    マハマド　ゴラム　ソフラブ, Makoto Miwa, Yutaka Sasaki

    The Association for Natural Language Processing  ( Tohoku University )   2016.3 

    　More details

    Presentation type：Oral presentation (general)  

    Overview of Text Classification
    Semantically Augmented Statistical Vector Space Model
    Word Embedding in Deep Learning
    Edge-based Learning in Deep Architechture
    Learning in Deep with Dual Coordinate Decent-SVM
    Evaluation
    Conclusion

  • GAによる構文木枝刈りを用いた単一文書要約

    立林裕太朗 知能数理*, 藤田充洋 豊田中研*, 三輪　誠, 古谷　克司, 佐々木　裕

    第22回言語処理学会年次大会  ( 東北大学 )   2016.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 文書・文間及びカテゴリ間の関係を考慮したレーティング予測

    外山洋太 知能数理*, 三輪　誠, 佐々木　裕

    第22回言語処理学会年次大会  ( 東北大学 )   2016.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • ニューラルキャプション生成モデルによる画像説明文の選択

    高里盛良 知能数理*, 三輪　誠, 佐々木　裕

    第22回言語処理学会年次大会  ( 東北大学 )   2016.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 英文穴埋め問題における文章ベクトルと学習データの質の影響

    森洸樹 知能数理*, 三輪　誠, 佐々木　裕

    第222回自然言語処理研究会  ( 首都大学東京秋葉原サテライトキャンパス )   2015.7  情報処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • Centroid-Means-Embedding: An Approach to Infusing Word Embeddings into Features for Text Classification International conference

    the 19th pacific-Asia Knowledge Discovery and Data Mining   ( Ho Chi Minh City, Vietnam )   2015.5  PAKDD Organizing Committee

    　More details

    Presentation type：Oral presentation (general)  

  • TTI'S Gender Prediction System using Bootstrapping and Identical-Hierarchy International conference

    the 19th pacific-Asia Knowledge Discovery and Data Mining   ( Ho Chi Minh City, Vietnam )   2015.5  PAKDD Organizing Committee

    　More details

    Presentation type：Oral presentation (general)  

  • 隠れ状態を用いたホテルレビューのレーティング予測

    藤谷宣典 知能数理*, 三輪　誠, 佐々木　裕

    第21回言語処理学会年次大会  ( 京都大学 )   2015.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 辞書と文脈情報を用いた対義語モデルの学習

    小野正貴 知能数理*, 三輪　誠, 佐々木　裕

    第21回言語処理学会年次大会  ( 京都大学 )   2015.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 語順と共起を考慮したニューラル言語モデルによる英文穴埋め

    森洸樹 知能数理*, 三輪　誠, 佐々木　裕

    第21回言語処理学会年次大会  ( 京都大学 )   2015.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • DCASVMを用いた高性能な大規模階層的文書分類

    佐々木　裕, マハマド　ゴラム　ソフラブ, 三輪　誠

    第21回言語処理学会年次大会  ( 京都大学 )   2015.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 文のカテゴリと極性の度合いの推定を行う評判分析システムの研究

    大竹翔馬 知能数理*, 三輪　誠, 佐々木　裕

    第21回言語処理学会年次大会  ( 京都大学 )   2015.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 交通オントロジーを対象とした質問文のSPARQLクエリ変換

    鈴木遼司 知能数理*, 三輪　誠, 佐々木　裕

    第21回言語処理学会年次大会  ( 京都大学 )   2015.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 交通オントロジーの半自動拡張のための交通用語認識

    河辺一仁 知能数理*, 三輪　誠, 佐々木　裕

    第21回言語処理学会年次大会  ( 京都大学 )   2015.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

  • 学習型質問応答における単語文脈ベクトルの効果の研究.

    牧瀬晃太 知能数理*, 三輪　誠, 佐々木　裕

    第21回言語処理学会年次大会  ( 京都大学 )   2015.3  言語処理学会

    　More details

    Presentation type：Oral presentation (general)  

▼display all

Awards

  • 第27回年次大会　Sansan　DSOC賞

    2021.3   言語処理学会  

    牧野晃平 知能数理*, 三輪　誠, 佐々木　裕

  • First prize in the n2c2 2019 Track 3

    2019.11   n2c2 2019 Challenge Organizing Committee   n2c2 2019 Track 3, Medical Entity Linkingにおいて世界各国から参加した33システム中１位．

    Tomoki Tsujimura*, Noriyuki Mori*, Masaki Asada*, Makoto Miwa, Yutaka Sasaki

  • 豊田奨学基金研究奨励賞

    2018.3   豊田工業大学  

    三輪　誠

  • Discovery Contest Honorable Mention

    2015.5   Pacific-Asia Knowledge Discovery and Data Mining   Participated PAKDD'15 Data Mining Competition: Gender Prediction Based on E-commerce Data Rank: 4th out of 149 participants

    マハマド　ゴラム　ソフラブ, Makoto Miwa, Yutaka Sasaki

  • 情報処理学会論文誌ジャーナル/JIP特選論文

    2014.11   情報処理学会  

    水上直紀 東京大学*, 中張遼太郎 東京大学*, 浦晃 東京大学*, 三輪　誠, 鶴岡慶雅 東京大学*, 近山隆 東京大学*

Grant-in-Aid for Scientific Research