@article{ART003339640},
author={jung chul and Soosang Lee and Rho Jee-Hyun},
title={A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation},
journal={Journal of the Korean Society for Library and Information Science},
issn={1225-598X},
year={2026},
volume={60},
number={2},
pages={387-408}
TY - JOUR
AU - jung chul
AU - Soosang Lee
AU - Rho Jee-Hyun
TI - A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation
JO - Journal of the Korean Society for Library and Information Science
PY - 2026
VL - 60
IS - 2
PB - 한국문헌정보학회
SP - 387
EP - 408
SN - 1225-598X
AB - The purpose of this study is to empirically analyze the effects of bibliographic element augmentation on the performance of automatic KDC classification model. To this end, bibliographic data were collected from the BookNet Korea (BNK) system, and three bibliographic element sets were constructed according to the level of bibliographic augmentation: Set A (title, author), Set B (title, author, keywords), and Set C (title, author, keywords, book summaries, tables of contents). KLUE-BERT-based automatic classification models were then applied to each set, and classification performance at the KDC division level was comparatively analyzed. The results showed that, compared with Model A, classification performance improved across all KDC main classes in Model C. In particular, the Arts (6XX) exhibited the highest rate of improvement, with a 124.24% increase in F1-score. However, depending on the level of bibliographic augmentation, some main classes showed performance degradation and no significant change. Additional analysis at the division level identified four patterns of performance change: consistently improved, conditionally improved, decreased, and unchanged. These findings suggest that improving the performance of automatic KDC classification requires a combined consideration of the structural characteristics of bibliographic elements and the subject characteristics of individual divisions.
KW - Automatic Classification;Korean Decimal Classification(KDC);Deep Learning;Bibliographic Element Augmentation;BERT Model
DO -
UR -
ER -
jung chul, Soosang Lee and Rho Jee-Hyun. (2026). A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation. Journal of the Korean Society for Library and Information Science, 60(2), 387-408.
jung chul, Soosang Lee and Rho Jee-Hyun. 2026, "A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation", Journal of the Korean Society for Library and Information Science, vol.60, no.2 pp.387-408.
jung chul, Soosang Lee, Rho Jee-Hyun "A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation" Journal of the Korean Society for Library and Information Science 60.2 pp.387-408 (2026) : 387.
jung chul, Soosang Lee, Rho Jee-Hyun. A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation. 2026; 60(2), 387-408.
jung chul, Soosang Lee and Rho Jee-Hyun. "A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation" Journal of the Korean Society for Library and Information Science 60, no.2 (2026) : 387-408.
jung chul; Soosang Lee; Rho Jee-Hyun. A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation. Journal of the Korean Society for Library and Information Science, 60(2), 387-408.
jung chul; Soosang Lee; Rho Jee-Hyun. A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation. Journal of the Korean Society for Library and Information Science. 2026; 60(2) 387-408.
jung chul, Soosang Lee, Rho Jee-Hyun. A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation. 2026; 60(2), 387-408.
jung chul, Soosang Lee and Rho Jee-Hyun. "A Study on the Performance Changes in Automatic KDC Classification through Bibliographic Element Augmentation" Journal of the Korean Society for Library and Information Science 60, no.2 (2026) : 387-408.
