증거-추론 혼합방법 기반 국가 R&D 모니터링 연구

박진한,권태훈,윤민영,임정선 한국유통물류정책학회 2024 유통물류연구 Vol.11 No.1

세계 주요국 정부들은 국가 차원의 R&D 투자와 정책 수립을 위해 실측 가능한 증거기반(Evidence-Based)정책지원 시스템을 연구개발 및 고도화하고 있으며, 정책 수립 의사결정자의 판단을 보조하기 위해 인공지능을 활용한 분석체계를 도입하고 있다. 한국 정부도 국가 R&D와 기술혁신을 위해 세계 주요국들의 증거 기반 정책지원 시스템의 분석체계와 유사한 연구를 진행하고 있다. 이 연구는 국가 R&D 성과 모니터링에 필요한 의사결정지원시스템을 개발하기 위한 목적으로 인공지능과 기계학습 및 해석 방법론인 증거-추론 혼합방법을 적용하고, 국가 단위의 R&D 자료가 있는 NTIS DB를활용하여, 해양수산부가 수행한 R&D 투자의 성과를 분석하였다. 분석 대상은 해양수산부가 수행한 R&D 과제의 사업화 성과로 발생한 2016∼2020년의 매출액 중 100억 원 이상의 매출액이 발생한 부산, 경남, 서울이며, 이 지역의 6T 분류에 해당하는 R&D 성과를 분석하였다. R&D 투자로 해당 기간 동안 발생한 전국 약 950억 원의 매출액 중에서 부산이 약 390억 원, 경남이 약 152억 원, 서울이 약 143억 원을 차지하는 것으로 나타났다. 부산에서 발생한 R&D 사업화에 의한 매출액은 ET 분야가 62%, 기타 기술 분야가 20% 수준으로 나타났으며, 전국 ET 분야 매출액의 75%가 부산에서 발생하였다. 경남에서 발생한 R&D 사업화에 의한 매출액은BT 분야가 147억 원으로 경남 전체의 96%를 차지하는 것으로 나타났으며, 전국 BT 분야 매출액의 36%가 경남에서 발생하였다. 서울에서 발생한 R&D 사업화에 의한 매출액은 BT 분야가 77억 원, ET 분야가 53.9억 원으로 나타났다. 본 연구의 학문적 시사점은 국가 단위의 R&D 투자와 사업화 성과를 실증 자료와 인공지능 기반의 방법론을 활용하여 추적과 설명 가능한 수준의 분석 결과를 도출하는 시도로서 의미가 있다. 실무적 시사점은 본 연구모형이 제안하는 분석 방법과 가공한 자료를 활용해 분석한 객관적 결과를 활용하여 국가 단위의 R&D 정책 수립 시 의사결정에 도움을 받을 수 있다는 점이다. Governments of major countries worldwide are actively researching and advancing evidence-based policy supportsystems for national-scale R&D investment and policy formulation. To assist policy decision-makers, these systemsincorporate artificial intelligence-driven analytical frameworks. The South Korean government, in alignment withglobal trends, is conducting research on evidence-based policy support systems of leading nations. This study, with the objective of developing a decision support system for national R&D monitoring, applies artificialintelligence, and the evidence-inference hybrid methodology. Using the National Science & Technology InformationService (NTIS) database containing national-level R&D data, the research analyzes the outcomes of R&D investmentsconducted by the Ministry of Oceans and Fisheries. The analysis focuses on the commercialization outcomes of R&Dprojects in the regions of Busan, Gyeongnam, and Seoul, which generated revenue exceeding 10 billion KRW between2016 and 2020 within the 6T classification. The results indicate that, out of the approximately 95 billion KRW in national revenue generated during the specifiedperiod, Busan accounted for approximately 39 billion KRW, Gyeongnam for approximately 15 billion KRW, and Seoulfor around 14 billion KRW. In Busan, R&D commercialization in the ET field contributed 62% of the revenue, withother technical fields contributing 20%. Notably, 75% of the national ET field revenue originated from Busan. Gyeongnam’s R&D commercialization, mainly in the BT field, accounted for 96% of the region’s total, representing36% of the national BT field revenue. In Seoul, R&D commercialization led to revenue of 7.7 billion KRW in the BTfield and 5.4 billion KRW in the ET field. From an academic perspective, this study contributes by attempting to derive analytically traceable and explainableresults of commercialization outcomes by national R&D investment, using evidence data with an artificial intelligence-based inference. From a practical standpoint, the proposed analysis method and processed data offerobjective results that can aid decision-making in national R&D policy formulation.

한약재 유통실태와 개선방향에 대한 연구

박진한 慶州大學校 2005 論文集 Vol.18 No.1

This thesis studied the current situations of production, infrastructure, distribution, consumption patterns of medicinal herbs through literature survey as well as the first hand knowledge of the industry. The use of medicinal herbs, that have been used mainly as medicine according to the principle of oriental medicine, are rapidly expanding as they are being used as cosmetics and dynamic foods, enabled by the development of the pharmaceutical and manufacturing technology. However, even with the rapid change in the consumption patterns of medicinal herbs, the production, distribution, and quality management of medicinal herbs are still way behind of the time. In conclusion, this study suggests recommendations to improve production, distribution, and quality control systems and practices of the medicinal herbs industry.

일축스크류압출기를 이용한 폴리카보네이트의 압출에 관한 연구

박진한,김용욱,류승훈 경희대학교 산학협력기술연구원 2000 산학협력기술연구논문집 Vol.6 No.-

Determination of the residence time distribution(RTD) of the polymer is very important on the product quality. Measurement of residence time distribution by the UV tracer were carried out during this study. The effect of screw speed, barrel temperature, molecular weight and molecular structure on the RTD was examined on three grades of polycarbonate. When the screw speed and barrel temperature were increased, just linear polycarbonate of high molecular weight was found to result in a broader distribution. Results obtained from the experiments were compared with Tadmor's melt extrusion model. It is clear that the linear polycarbonate of high molecular weight better agreement between Tadmor's melt extrusion model and experimental datum.

경주 특화작물 체리의 항산화 효능

박진한 경주지역발전협의회 부설 경주발전연구원 2012 경주연구 Vol.21 No.2

Multi-person Multi-attribute Decision Making Problems Based on Interval-valued Intuitionistic Fuzzy Information

박진한,권영철,손미정 한국지능시스템학회 2010 INTERNATIONAL JOURNAL of FUZZY LOGIC and INTELLIGE Vol.10 No.4

Based on the interval-valued intuitionistic fuzzy hybrid geometric (IIFHG) operator and the interval-valued intuitionistic fuzzy weighted geometric (IIFWG) operator, we investigate the group decision making problems in which all the information provided by the decision-makers is presented as interval-valued intuitionistic fuzzy decision matrices where each of the elements is characterized by interval-valued intuitionistic fuzzy numbers, and the information about attribute weights is partially known. A numerical example is used to illustrate the applicability of the proposed approach.

An approach based on the LOWHM and induced LOWHM operators to group decision making under linguistic information

박진한,이부영,손미정 한국지능시스템학회 2010 한국지능시스템학회논문지 Vol.20 No.2

In thispaper,weproposeaninducedlinguisticorderedweightedharmonicmean(ILOWHM)operator.TheILOWHM operator ismoregeneraltypeofaggregationoperator,whichisbasedontheLHMandLOWHMoperators.Basedonthe LOWHMandILOWHMoperator,wedevelopanapproachtogroupdecisionmakingwithlinguisticpreferencerelations. Finally,anapplicationoftheapproachtogroupdecisionmakingproblemwithlinguisticpreferencerelationsispointed out.

Papillary Fibroelastoma Presenting as a Left Ventricular Mass

박진한,설상훈,조환진,박시형,김동기,김웅,양태현,김대경,김두일,김동수 한국심초음파학회 2010 Journal of Cardiovascular Imaging (J Cardiovasc Im Vol.18 No.2

Cardiac papillary fibroelastoma (CPF) is a benign cardiac tumor that usually affects cardiac valves. It is usually discovered incidentally on routine echocardiography. However, left ventricular CPF is rare. This report describes the case of a 73-year-old female, referred to a cardiology department for evaluation of a mass of the left ventricle. The mass was found routine echocardiography. The transthoracic echocardiography revealed a 2.2×1.3 cm highly oscillating mass, attached by stalk on the inferior wall of the left ventricle. Cardiac magnetic resonance imaging demonstrated a non-enhanced, 1.8×1.0 cm mass on the inferior wall of the left ventricle. The patient underwent surgical resection of the mass, histopathologic examination of the mass confirmed the diagnosis of a CPF.

일본 근세 도시사 연구의 전개와 과제

박진한 연세대학교 국학연구원 2007 동방학지 Vol.139 No.-

An approach based on the induced LOWHM operators to group decision making

박진한,박용범,이부영,손미정 한국지능시스템학회 2010 한국지능시스템학회 학술발표 논문집 Vol.20 No.1

In this paper, we propose an induced linguistic ordered weighted harmonic mean (ILOWHM) operator. The ILOWHM operator is more general type of aggregation operator, which is based on the LHM and LOWHM operators. Based on the LOWHM and ILOWHM operator, we develop an approach to group decision making with linguistic preference relations. Finally, an application of the approach to group decision making problem with linguistic preference relations is pointed out.

Dynamic linguistic multiple attribute decision making

박진한,박용범,조현주 한국지능시스템학회 2010 한국지능시스템학회 학술발표 논문집 Vol.20 No.2

The notion of dynamic linguistic variable is are defined, and new aggregation operator such as the dynamic linguistic weighted harmonic mean (DLWHM) operator, which can be utilized to aggregate preference information taking the form of linguistic variables, and then study some desirable properties of the operator. A procedure based on the DLWHM operator is developed to solve the dynamic linguistic multiple attribute decision making (DLMADM) problems where all decision information about values takes the form of linguistic variables collected at different periods. Finally, an illustrative example is given to verify the developed approaches and to demonstrate their practicality and effectiveness.