TY - JOUR
T1 - Multi-purpose technology commercialization recommender system with large-scale Korean language model
AU - Noh, Kangjun
AU - Hwang, Hyeji
AU - Lim, Yongtaek
AU - Oh, Changdae
AU - Kim, Seungyeon
AU - Lee, Eunkyeong
AU - Choi, Yunjeong
AU - Kim, Sungjin
AU - Choi, Hosik
AU - Song, Kyungwoo
N1 - Publisher Copyright:
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.
PY - 2025/6
Y1 - 2025/6
N2 - Abstract: Large Language Models (LLMs) have demonstrated significant performance improvements and are widely applied across various domains. However, their application in technology commercialization (TC) remains largely unexplored. Effectively utilizing LLMs for TC requires a deep understanding of TC-related documents, which is crucial for accurately classifying relevant sentences and recommending documents that align with consumers’ needs. To address this, we construct a diversified TC-related tagging dataset by defining four new tagging structures and collecting 34,298 Korean tagged sentences. Additionally, we propose a novel multi-purpose TC recommendation algorithm that considers the diverse objectives of consumers, ensuring a more flexible and practical recommendation system. Graphic abstract: (Figure presented.)
AB - Abstract: Large Language Models (LLMs) have demonstrated significant performance improvements and are widely applied across various domains. However, their application in technology commercialization (TC) remains largely unexplored. Effectively utilizing LLMs for TC requires a deep understanding of TC-related documents, which is crucial for accurately classifying relevant sentences and recommending documents that align with consumers’ needs. To address this, we construct a diversified TC-related tagging dataset by defining four new tagging structures and collecting 34,298 Korean tagged sentences. Additionally, we propose a novel multi-purpose TC recommendation algorithm that considers the diverse objectives of consumers, ensuring a more flexible and practical recommendation system. Graphic abstract: (Figure presented.)
KW - BERT
KW - LM
KW - Machine learning
KW - Multi-purpose recommendation
KW - Technology commercialization
UR - http://www.scopus.com/inward/record.url?scp=105006977006&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=105006977006&partnerID=8YFLogxK
U2 - 10.1007/s11227-025-07341-4
DO - 10.1007/s11227-025-07341-4
M3 - Article
AN - SCOPUS:105006977006
SN - 0920-8542
VL - 81
JO - Journal of Supercomputing
JF - Journal of Supercomputing
IS - 8
M1 - 942
ER -