Large Language Model-based R&D Solution Analysis Approach Using Problem-Solution Information of Patents
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 l...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Korea Institute of Intellectual Property
2024-09-01
|
| Series: | Journal of Intellectual Property |
| Subjects: | |
| Online Access: | https://jip.or.kr/1903-08/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849312978366627840 |
|---|---|
| author | Seunghyun Lee Jiho Lee Seoin Park Jae-Min Lee Hong-Woo Chun Janghyeok Yoon |
| author_facet | Seunghyun Lee Jiho Lee Seoin Park Jae-Min Lee Hong-Woo Chun Janghyeok Yoon |
| author_sort | Seunghyun Lee |
| collection | DOAJ |
| description | 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. |
| format | Article |
| id | doaj-art-cc2b52bc8cf840ca9bf35381e0e50494 |
| institution | Kabale University |
| issn | 1975-5945 2733-8487 |
| language | English |
| publishDate | 2024-09-01 |
| publisher | Korea Institute of Intellectual Property |
| record_format | Article |
| series | Journal of Intellectual Property |
| spelling | doaj-art-cc2b52bc8cf840ca9bf35381e0e504942025-08-20T03:52:52ZengKorea Institute of Intellectual PropertyJournal of Intellectual Property1975-59452733-84872024-09-0119315518010.34122/jip.2024.19.3.155Large Language Model-based R&D Solution Analysis Approach Using Problem-Solution Information of PatentsSeunghyun Lee0Jiho Lee1Seoin Park2Jae-Min Lee3Hong-Woo Chun4Janghyeok Yoon5https://orcid.org/0000-0002-8701-0695PhD Candidate, Department of Industrial Engineering, Konkuk University, Republic of KoreaDirector, AI Lab, Neopons Inc., Republic of KoreaMaster’s Student, Department of Industrial Engineering, Konkuk University, Republic of KoreaPrincipal Researcher, Future Technology Analysis Center, Korea Institute of Science and Technology Information, Republic of KoreaDirector, Future Technology Analysis Center, Korea Institute of Science and Technology Information, Republic of KoreaProfessor, Department of Industrial Engineering, Konkuk University, Republic of KoreaPatents, 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.https://jip.or.kr/1903-08/patent analysisproblem-solution informationr&d solutionlarge language modelfine-tuning |
| spellingShingle | Seunghyun Lee Jiho Lee Seoin Park Jae-Min Lee Hong-Woo Chun Janghyeok Yoon Large Language Model-based R&D Solution Analysis Approach Using Problem-Solution Information of Patents Journal of Intellectual Property patent analysis problem-solution information r&d solution large language model fine-tuning |
| title | Large Language Model-based R&D Solution Analysis Approach Using Problem-Solution Information of Patents |
| title_full | Large Language Model-based R&D Solution Analysis Approach Using Problem-Solution Information of Patents |
| title_fullStr | Large Language Model-based R&D Solution Analysis Approach Using Problem-Solution Information of Patents |
| title_full_unstemmed | Large Language Model-based R&D Solution Analysis Approach Using Problem-Solution Information of Patents |
| title_short | Large Language Model-based R&D Solution Analysis Approach Using Problem-Solution Information of Patents |
| title_sort | large language model based r d solution analysis approach using problem solution information of patents |
| topic | patent analysis problem-solution information r&d solution large language model fine-tuning |
| url | https://jip.or.kr/1903-08/ |
| work_keys_str_mv | AT seunghyunlee largelanguagemodelbasedrdsolutionanalysisapproachusingproblemsolutioninformationofpatents AT jiholee largelanguagemodelbasedrdsolutionanalysisapproachusingproblemsolutioninformationofpatents AT seoinpark largelanguagemodelbasedrdsolutionanalysisapproachusingproblemsolutioninformationofpatents AT jaeminlee largelanguagemodelbasedrdsolutionanalysisapproachusingproblemsolutioninformationofpatents AT hongwoochun largelanguagemodelbasedrdsolutionanalysisapproachusingproblemsolutioninformationofpatents AT janghyeokyoon largelanguagemodelbasedrdsolutionanalysisapproachusingproblemsolutioninformationofpatents |