http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
데이터 마이닝을 활용한 한국 프로야구 구단의 승패예측과 승률 향상을 위한 전략 도출 연구
김원종 ( Kim Won-jong ),최연식 ( Choi Yeon-sik ),유동희 ( Yoo Dong-hee ) 한국스포츠산업경영학회 2018 한국스포츠산업경영학회지 Vol.23 No.3
본 연구에서는 데이터 마이닝 기법을 활용하여 한국 프로야구의 승패예측모형을 구축하는 실험을 진행하였다. 이를 위해, 2017년에 실시된 한국 프로야구 10개 구단의 전체 경기 중 무승부 경기를 제외한 1,418경기에 대한 자료를 사용하였다. 승패예측모형에는 의사결정나무, 베이즈넷, 인공신경망 알고리즘과 앙상블 기법인 배깅과 부스팅이 사용되었으며, 그 결과 배깅 기법에 인공신경망을 적용한 예측모형에서 가장 높은 예측률인 85.18%를 기록하였다. 다음으로 의사결정나무 기반 예측모형을 활용하여 한국 프로야구 전체 구단에 관한 8개의 승패규칙을 도출하였다. 여기에서 승패규칙은 승패예측에 영향을 미치는 주요 요인들인 팀출루율, 팀타율, 피안타, 안타, 타석, 타수로 표현되며, 도출된 규칙을 바탕으로 구단의 승률 향상에 도움을 주는 전략을 제안하였다. 또한 플레이오프 진출 구단과 미진출 구단에 관한 승리규칙을 각각 4개씩 도출하였고 이를 바탕으로 두 집단에 맞춤화된 승률 향상 전략을 제시하였으며, 실제 구단에서 선수를 영입한 방향과의 비교를 통해 제시된 전략의 활용 가능성을 확인하였다. This study conducted an experiment to develop win-loss prediction models for the Korean professional baseball league using data mining techniques. To this end, we used data on 1,418 games from all games played by the ten Korean professional baseball teams in 2017, except draw games. We developed win-loss prediction models using not only a decision tree, Bayse net, and artificial neural network algorithms, but also ensemble methods, such as bagging and boosting. As a result, first, we found that the artificial neural network-based prediction model using the bagging method reported the best accuracy (85.18%). Second, we derived eight win-loss rules for entire teams from the decision tree-based prediction model. These rules consist of six influential factors: team on base average, team batting average, hit by opponent, hit, plate appearances, and at bat. Using the derived rules, we proposed helpful strategies for improving the winning rate. Third, we derived four winning rules for both playoff teams and non-playoff teams; using the rules we proposed customized strategies for improving the winning rate of the two different groups. Finally, we confirmed the feasibility of the proposed strategies by comparing non-playoff teams’ actual player recruitment strategies.
Preference Analysis for Flower Jewelry Using Conjoint Analysis
김연희(Yeon Hee Kim),김미진(Mi Jin Kim),윤숙영(Suk Young Yun),장태원(Taeg Won Chang),최병진(Byung Jin Choi) 인간식물환경학회 2022 인간식물환경학회지 Vol.25 No.3
Background and objective: This study was conducted to determine the preferences in flower jewelry by conducting a surveyusing conjoint analysis on the perception of flower jewelry and provide basic data for development of flower jewelryproducts. Methods: For the conjoint analysis, four attributes of flower jewelry were selected. Attribute 1 was related to the mainmaterial of flower jewelry, presented in three levels: cut flowers, potted flowers, and processed flowers. Attribute 2 wasrelated to wearing area presented in five levels of wearing area: hair, neck, ear, arm (wrist) and chest (shoulder). Attribute3 was related to price provided for selling products in flower shops, presented in three levels: less than 30,000 won,50,000-70,000 won, and more than 100,000 won. Attribute 4 was related to flower jewelry wearing time, presented inthree levels: less than 6 hours, 12-24 hours, and more than 24 hours. After extracting 25 profiles through orthogonaldesign, a questionnaire survey was conducted on 402 general participants, from which the responses of 343 participantswere analyzed. Results: By attribute of flower jewelry, the respondents attached most importance to price (34.14%), followed by wearingarea (29.18%), wearing time (18.51%), and material (18.17%), proving that the general public attached most importanceto price (Pearson's R = .987, p = .000). The preference in levels by attribute was highest for processed flowers in material,less than 30,000 won in price, hair in wearing area, and 12-24 hours in wearing time. Conclusion: By making flower jewelry based on the results of the analysis, it would be possible to help increase items soldin flower shops as well as their income.
Hee-Sang Park,Man-Yong Choi,Koo-Ahn Kwon,Jeong-Hak Park,Won-Jae Choi,Hyun-Chul Jung 한국비파괴검사학회 2017 한국비파괴검사학회지 Vol.37 No.2
Recently, composite materials have been mainly used in the main wings, ailerons, and fuselages of aircraft and rotor blades of helicopters. Composite materials used in rapid moving structures are subject to impact by hail, lightning, and bird strike. Such an impact can destroy fiber tissues in the composite materials as well as deform the composite materials, resulting in various problems such as weakened rigidity of the composite structure and penetration of water into tiny cracks. In this study, experiments were conducted using a 2 kW halogen lamp which is most frequently used as a light source, a 2 kW near-infrared lamp, which is used for heating to a high temperature, and a 6 kW xenon flash lamp which emits a large amount of energy for a moment. CFRP composite sandwich panels using Nomex honeycomb core were used as the specimens. Experiments were carried out under impact damages of 1, 4 and 8 J. It was found that the detection of defects was fast when the xenon flash lamp was used. The detection of damaged regions was excellent when the halogen lamp was used. Furthermore, the near-infrared lamp is an effective technology for showing the surface of a test object.
[P344] A case of squamous cell carcinoma on the finger induced by radiation
( Hee Seong Yoon ),( Ji Hye Heo ),( Si Hyub Lee ),( Seung Dohn Yeom ),( Jeonghyun Shin ),( Gwang Seong Choi ),( Ji Won Byun ) 대한피부과학회 2017 대한피부과학회 학술발표대회집 Vol.69 No.1
Squamous cell carcinoma(SCC) of the nail is most frequent malignant tumor of the nail unit. When SCC occurs in the finger nail, it occurs particularly in thumb and index finger. Etiologic factors of SCC of the nail unit include chronic trauma, human papilloma virus infection and radiodermatitis. We report a case of a 48-year-old man with painful crusted necrotic plaques on the right thumb, index, third finger and left thumb, index finger two years ago. The patient was an orthopedic surgeon and he was repeatedly exposed to radiation due to frequent use of C-arm for 11 years. During the 11-year period, He performed an average of 700 cases of percutaneous vertebral augumentation, 70,000 cases of nerve block per year. Skin biopsy on right index finger, right thumb, left thumb showed that atypical epidermal cells proliferate downward into the dermis. On right hand MRI, bone invasion was found in right thumb and right index finger. We diagnosed as SCC in multidisciplinary cancer care. He refused amputation. For the right thumb and index finger, SCC lesion was widely excised and and distal phalange of index finger which had a high possibility of bone involvement was partially severed leaving 6mm and the defect was repaired with full thickness skin graft. Mohs micrographic surgery and skin graft was performed on left thumb. All fingers suspected of the skin cancer in situ due to chronic radiation dermatitis were treated with ingenol mebutate.
Estimates of the Number of Workers Exposed to Diesel Engine Exhaust in South Korea from 1993 to 2013
Choi, Sangjun,Park, Donguk,Kim, Seung Won,Ha, Kwonchul,Jung, Hyejung,Yi, Gwangyong,Koh, Dong-Hee,Park, Deokmook,Sun, Oknam,Uuksulainen, Sanni Occupational Safety and Health Research Institute 2016 Safety and health at work Vol.7 No.4
Background: The aim of this study was to estimate the number of workers exposed to diesel engine exhaust (DEE) by industry and year in the Republic of Korea. Method: The estimates of workers potentially exposed to DEE in the Republic of Korea were calculated by industry on the basis of the carcinogen exposure (CAREX) surveillance system. The data on the labor force employed in DEE exposure industries were obtained from the Census on Establishments conducted by the Korea National Statistical Office from 1993 to 2013. The mean values of prevalence rates adopted by EU15 countries were used as the primary exposure prevalence rates. We also investigated the exposure prevalence rates and exposure characteristics of DEE in 359 workplaces representing 11 industries. Results: The total number of workers exposed to DEE were estimated as 270,014 in 1993 and 417,034 in 2013 (2.2% of the total labor force). As of 2013, the industry categorized as "Land transport" showed the highest number of workers exposed to DEE with 174,359, followed by "Personal and household services" with 70,298, "Construction" with 45,555, "Wholesale and retail trade and restaurants and hotels" with 44,005, and "Sanitation and similar services" with 12,584. These five industries, with more than 10,000 workers exposed to DEE, accounted for 83% of the total DEE-exposed workers. Comparing primary prevalence rates used for preliminary estimation among 49 industries, "Metal ore mining" had the highest rate at 52.6%, followed by "Other mining" with 50.0%, and "Land transport" with 23.6%. Conclusion: The DEE prevalence rates we surveyed (1.3-19.8%) were higher than the primary prevalence rates. The most common emission sources of DEE were diesel engine vehicles such as forklifts, trucks, and vans. Our estimated numbers of workers exposed to DEE can be used to identify industries with workers requiring protection from potential exposure to DEE in the Republic of Korea.