BERT 모형을 이용한 주제명 자동 분류 연구 (A Study on Automatic Classification of Subject Headings Using BERT Model)

이용구 (LeeYong-Gu)

doi:10.4275/KSLIS.2023.57.2.435

A Study on Automatic Classification of Subject Headings Using BERT Model

Journal of the Korean Society for Library and Information Science
2023, 57(2), pp.435-452
DOI : 10.4275/KSLIS.2023.57.2.435
Publisher : 한국문헌정보학회
Research Area : Interdisciplinary Studies > Library and Information Science
Received : April 21, 2023
Accepted : May 19, 2023
Published : May 31, 2023

LeeYong-Gu ¹

¹경북대학교

Excellent Accredited

ABSTRACT

This study experimented with automatic classification of subject headings using BERT-based transfer learning model, and analyzed its performance. This study analyzed the classification performance according to the main class of KDC classification and the category type of subject headings. Six datasets were constructed from Korean national bibliographies based on the frequency of the assignments of subject headings, and titles were used as classification features. As a result, classification performance showed values of 0.6059 and 0.5626 on the micro F1 and macro F1 score, respectively, in the dataset (1,539,076 records) containing 3,506 subject headings. In addition, classification performance by the main class of KDC classification showed good performance in the class General works, Natural science, Technology and Language, and low performance in Religion and Arts. As for the performance by the category type of the subject headings, the categories of plant, legal name and product name showed high performance, whereas national treasure/treasure category showed low performance. In a large dataset, the ratio of subject headings that cannot be assigned increases, resulting in a decrease in final performance, and improvement is needed to increase classification performance for low-frequency subject headings.

KEYWORDS

Automatic Classification, Deep Learning, BERT Model, Automated Subject Indexing, Automated Subject Classification

Citation status

* References for papers published after 2023 are currently being built.

[web] 국립중앙도서관 / 2021 / 인공지능이 서비스하는 검색과 요약, 체험해 보세요 / https://www.nl.go.kr/NL/contents/N50603000000.do?schM=view&id=38537&schBcid=normal0302

[journal] 김인후 / 2022 / Automatic Classification of Academic Articles Using BERT Model Based on Deep Learning / 정보관리학회지 / 한국정보관리학회 39(3) : 293~310

[journal] 박진우 / 2022 / A Study on Automatic CPC Classification based on Korean Patent Sentence ―A Deep Learning Approach using Artificial Intelligence Language Model KorPatBERT― / 지식재산연구 / 한국지식재산연구원 17(3) : 209~256

[journal] 백지원 / 2014 / A Study on Improving Access & Retrieval System of the National Library of Korea Subject Headings / 정보관리학회지 / 한국정보관리학회 31(1) : 31~51

[journal] 엄기홍 / 2021 / Automated Classification Model for Online Public Opinions in a Political Arena: KoBERT based Sentiment Analysis / 한국정당학회보 / 한국정당학회 20(3) : 167~191

[report] 오원석 / 2021 / 인공지능 기술을 활용한 사서 업무 지원 도구 개발에 관한 연구(979-11-6513-187-6) / 국립중앙도서관

[journal] 최윤경 / 2014 / A Study on Improvements for High Quality in National Library of Korea Subject Headings List / 한국문헌정보학회지 / 한국문헌정보학회 48(1) : 75~95

[journal] 황상흠 / 2020 / BERT-based Classification Model for Korean Documents / 한국전자거래학회지 / 한국전자거래학회 25(1) : 203~214

[other] Cho, K. / 2014 / Learning phrase representations using RNN encoder-decoder for statistical machine translation / arXiv preprint arXiv:1406.1078

[other] Devlin, J. / 2018 / BERT: pre-training of Deep Bidirectional Transformers for Language Understanding / arXiv preprint arXiv:1810.04805

[journal] Hochreiter, S. / 1997 / Long short-term memory / Neural Computation 9(8) : 1735~1780

[confproc] Jin, Q. / 2018 / AttentionMeSH : simple, effective and interpretable automatic MeSH indexer / Proceedings of the 6th BioASQ Workshop A challenge on large-scale biomedical semantic indexing and question answering : 47~56

[confproc] Kim, Y. / 2014 / Convolutional Neural Networks for Sentence Classification / Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing : 1746~1751

[confproc] LeCun, Y. / 1998 / Gradient-based learning applied to document recognition / Proceedings of the IEEE 86(11) : 2278~2324

[confproc] Mikolov, T. / 2011 / Extensions of recurrent neural network language model / 2011 IEEE international conference on acoustics, speech and signal processing / IEEE : 5528~5531

[other] Mork, J. G. / 2013 / The NLM Medical Text Indexer System for Indexing Biomedical Literature / BioASQ@ CLEF

[journal] Peng, S. / 2016 / Deepmesh: deep semantic representation for improving large-scale MeSH indexing / Bioinformatics 32(12) : i70~

[journal] Reich, P. / 1991 / Indexing Consistency : the Input/Output Function of Thesauri / College & Research Libraries 52(4) : 336~342

[journal] Saarti, J. / 2002 / Consistency of subject indexing of novels by public library professionals and patrons / Journal of Documentation 58(1) : 49~65

[journal] Tonta, Y. / 1991 / A study of indexing consistency between library of congress and british library catalogers / Library Resources & Technical Services 35(2) : 177~185

[journal] Vaswani, A. / 2017 / Attention Is All You Need / Advances in neural information processing systems 30 : 6000~6010

[journal] You, R. / 2021 / BERTMeSH : deep contextual representation learning for large-scale high-performance MeSH indexing with full text / Bioinformatics 37(5) : 684~692

KJCKorea
Journal Central

Journal of the Korean Society for Library and Information Science 2023 KCI Impact Factor : 0.84

A Study on Automatic Classification of Subject Headings Using BERT Model

ABSTRACT

KEYWORDS

Citation status

* References for papers published after 2023 are currently being built.

Journal of the Korean Society for Library and Information Science 2023 KCI Impact Factor : 0.84

A Study on Automatic Classification of Subject Headings Using BERT Model

ABSTRACT

KEYWORDS

Statistics

Tools

Issue List

Citation status

KCI Citation Counts (0)

REFERENCES (22) * References for papers published after 2023 are currently being built.

Search PDF

Citation

* References for papers published after 2023 are currently being built.