특허의 Problem-Solution 정보를 활용한 대규모 언어모델 기반 R&D 솔루션 분석 방법
Large Language Model-based R&D Solution Analysis Approach Using Problem-Solution Information of Patents
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
문제 해결을 위한 연구개발(R&D) 활동의 산출물인 특허는 Problem-Solution 정보의 집약체로 간주된다. 문제에 대한 해결방법을 탐색하기 위해 Problem-Solution 기반의 특허분석 연구가 다수 수행되었지만, 대규모 언어모델을 활용하기 위한 시도는 미흡한 실정이다. 텍스트 요약 및 생성 등 자연어 처리작업에 효과적인 대규모 언어모델을 통해 특허의 Problem-Solution 정보를 학습한다면, 이미 존재하는 R&D 솔루션을 탐색하는 것에서 나아가 주어진 문제에 적용가능한 새로운 솔루션을 생성할 수 있다. 따라서, 본 연구는 대규모 언어모델을 활용하여 새로운 R&D 솔루션을 생성 및 분석하는 방법을 제시한다. 구체적으로 1) 대량의 특허를 수집하여 데이터베이스를 구축하고, 2) 특허의 공통출원서식 텍스트로부터 Problem-Solution 정보를 추출 및 데이터셋을 구축한 후, 3) Problem-Solution 데이터셋을활용하여 대규모 언어모델을 파인튜닝하고 R&D 솔루션을 생성한 다음, 4) R&D 솔루션을 분석하여기술 컨셉 포트폴리오 맵을 제시한다. 본 연구는 기존의 R&D 솔루션 탐색을 넘어 대규모 언어모델을활용하여 솔루션을 생성하고 기술 컨셉을 제시하는 새로운 방법을 제시한다. 따라서 본 연구는 R&D 분야의 문제 해결 프로세스에서 선택의 폭을 넓히고, 언어모델 기반의 R&D 혁신을 촉진하는 데 기여할 수 있다.
Abstract
Patents, i.e., the output of research and development (R&D) activities, are regarded as a concentration of Problem–Solution information. Despite various patent analysis studies aimed at solving problems, large language model (LLM)-based studies are scarce. LLMs, which are effective for natural language processing tasks, such as text summarization and generation, have been applied in numerous fields, including healthcare, finance, and law. By learning the Problem-Solution information of patents as an LLM instead of merely examining existing R&D solutions, one can generate new solutions applicable to a specified problem. Therefore, this study proposes an approach to generate and analyze new R&D solutions using LLMs. Our systematic approach involves 1) collecting numerous patents and constructing a database; 2) extracting Problem-Solution information from the Common Application Form section of patents and constructing a Problem-Solution dataset; 3) fine-tuning an LLM using the problem-solution dataset and generating R&D solutions; and 4) analyzing R&D solutions to present a technology concept portfolio map. This study extends beyond the existing R&D solution exploration, presents a new approach for generating solutions, and suggests technology concepts using LLMs. Therefore, this study contributes to the expansion of the available options and fosters innovation in R&D field.
참고문헌 (37)
- Semyon D. Savransky, "Engineering of creativity : Introduction to TRIZ methodology of inventive problem solving" (2000), CRC press
- 윤장혁, "SAO 기반의 의미론적 특허 유사성을 활용한 특허맵 생성방법", Entrue Journal of Informa tion Technology, 10(1) (2011)
- 이지호, "특허의 Problem-Solution 텍스트 마이닝을 활용한 기술경쟁정보 분석 방법", 지식재산연구, 13(3) (2018)
- 이치훈, "사전 학습된 한국어 BERT 의 전이학습을 통한 한국어 기계독해 성능개선에 관한 연구", 한국IT 서비스학회지, 19(5) (2020)
- 정재민, "비즈니스 기회 발굴을 위한 문제-해결방법 기반의 특허분석 방법", 지식재산연구, 15(2) (2020)
- Dave Van Veen, "Adapted large language models can outperform medical experts in clinical text summarization", Nature Medicine, 30(4) (2024)
- Georg Richter, "The impact of metadata on the accuracy of automated patent classification", World Patent Information, 27(1) (2005)
- Hyunseok Park, "Identifying patent infringement using SAO based semantic technological similarities", Scientometrics, 90(2) (2012)
- Jaewoong Choi, "Technology opportunity discovery under the dynamic change of focus techn ology fields : Application of sequential pattern mining to patent classifications", Technological Forecasting and Social Change, 148 (2019)
- Janghyeok Yoon, "Identifying rapidly evolving technological trends for R&D planni ng using SAO-based semantic patent networks", Scientometrics, 88(1) (2011)
- Janghyeok Yoon, "Detecting signals of new technological opportunities using semantic patent analysis and outlier detection", Scientometrics, 90(2) (2012)
- Jiho Lee, "An approach for discovering firm-specific technology opportunities : Application of link prediction to F-term networks", Technological Forecasting and Social Change, 168 (2021)
- Johan Schot, "Three frames for innovation policy : R&D, systems of innovation and transformative change", Research policy, 47(9) (2018)
- Jonathan H, Choi, "ChatGPT goes to law school", Journal of Legal Education, 71(3) (2021)
- Kyuwoong Kim, "Investigating technology opportunities : The use of SAOx analysis", Scientometrics, 118 (2019)
- Martin G. Moehrle, "Patent-based inventor profiles as a basis for human resource decisions in research and development", R&D Management, 35(5) (2005)
- Saika Wong, "Construction contract risk identification based on knowledge-augmented lang uage models", Computers in Industry, 157 (2024)
- Sungchul Choi, "An SAO-based text-mining approach for technology roadmapping using patent information", R&D Management, 43(1) (2013)
- Sunhye Kim, "Patent infringement analysis using a text mining technique based on SAO structure", Computers in Industry, 125 (2021)
- Xuefeng Wang, "Identifying R&D partners for dye-sensitized solar cells : a multi-level patent portfolio-based approach", Technology Analysis & Strategic Management, 31(3) (2019)
- Xuefeng Wang, "Measuring patent similarity with SAO semantic analysis", Scientometrics, 121 (2019)
- Yi Zhang, "How to combine term clumping and technology roadmapping for newly emerging science & technology competitive intelligence:“problem & solution” pattern based semantic TRIZ tool and case study", Scientometrics, 101 (2014)
- Youngjin Seol, "Towards firm-specific technology opportunities : A rule-based machine learn ing approach to technology portfolio analysis", Journal of Informetrics, 17(4) (2023)
- Yu Gu, "Domain-specific language model pretraining for biomedical natural language proces sing", ACM Transactions on Computing for Healthcare, 3(1) (2021)
- Yuen-Hsien Tseng, "Text mining techniques for patent analysis", Information processing & management, 43(5) (2007)
- Zhe Zheng, "Pretrained domain-specific language model for natural language processing tasks in the AEC domain", Computers in Industry, 142 (2022)
- "Language models are unsupervised multitask learners"[URL]
- 제3차 과학기술 기반 사회문제해결 종합계획, "과학기술정보통신부", 과학기술정보통신부, ~ (2023)
- Eunjeong L. Park, "KoNLPy: Korean natural language processing in Python", Proceedings of 26th Annual Conference on Human and Cognitive Language Technology (2014), Special Interest Group of Human and Cognitive Language Technology
- Youngho Kim, "Automatic discovery of technology trends from patent text", Proceedings of the 2009 ACM symposium on Applied Computing (2009), Association for Computing Machinery