http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
글루타르알데히드 가교 조건에 따른 콜라겐 섬유의 방사성 및 역학적 특성 분석
안현철,공다정,이현호,서주연,김태겸,이소희,여상영 한국섬유공학회 2020 한국섬유공학회지 Vol.57 No.6
Collagen is a natural protein with excellent biocompatibility. Purified collagen isused in various medical and cosmetic products. Fibers made from collagen are particularlyused for medical purposes, but their mechanical properties are not good; therefore, manufacturingprocesses such as crosslinking and blending are being researched and developedto reinforce fibers and improve fiber spinnability. Particularly, in the crosslinking process,fiber properties can be manipulated and adjusted based on the crosslinking agent. In thisstudy, the spinnability and mechanical properties of the fibers were analyzed according tothe crosslinking concentration and crosslinking process time. Through this, the crosslinkedcollagen spinning conditions that can achieve a sufficient crosslinking effect with a minimumconcentration and process time were examined. The processing efficiency andmechanical properties of the collagen fibers were improved by optimizing the crosslinkingconditions.
안현철,조현서,강호철,이정환 진주산업대학교 농업기술연구소 2001 農業技術硏究所報 Vol.14 No.-
This study was focused on fragmentation of habitat by the human activities which is road construct, forest road, cities, industry, rail road, a power cable tower, and forest fire etc. in the Gyeongsangnam-do Province. Forests provide the basic habitat for wildlife, which may be specific forest type such as a broad leaf or mixed forest. Fragmentation of forest stands is creating concern among persons involved with wildlife. Agricultural and urban developments have sometimes disruped contiguous stands of forest into isolated patch. The road and developments in the forest area adjacent to the habitat have broken up into a few patches. Ecobridge project connect the fragmentation of habitat and allow wildlife to access food, water and move among habitat patches in disturbed area. In the Gyeongsangnam-do Province be having to 76 path road in which 6 highway route, 22 national road and 48 province road, respectively. On fragmentation of habitat by the road, highway, nat'l and province road were disrupted in 36, 62 and 57 sites , respectively. Among them 16, 29 and 30 sites were necessitates a eocbridge construction highway, nat'l and province road, respectively The goal of the Eco-bridge project is to maintain a nearly contiguous green network for wildlife. Some developments in the habitats are divided because of existing roadways, dam, cities, farm land, recreation and leisure town, etc. and environmental issues which are interposed between the habitats.
Milticlass SVM model with order information
안현철,김경재 한국지능시스템학회 2006 INTERNATIONAL JOURNAL of FUZZY LOGIC and INTELLIGE Vol.6 No.4
Original Support Vsctor Machines (SVMs) by Vapnik were used for binary classification problems. Some researchers have tried to extend original SVM to multiclass classification. However, their studies have only focused on classifying samples into nominal categories. This study proposes a novel multiclass SVM model in order to handle ordinal multiple classes. Our suggested model may use less classifiers but predict more accurately because it utilizes additional hidden information, the order of the classes. To validate our model, we apply it to the real-world bond rating case. In this study, we compare the results of the model to those of statistical and typical machine learning techniques, and another multiclass SVM algorithm. The result shows that proposed model may improve classification performance in comparison to other typical multiclass classification algorithms.
심정지 후 치료적 저체온 요법을 시행한 환자의 사망예측에 있어 B-type natriuretic peptide 수치의 유용성
안현철,조진성,김진주,양혁준,박원빈,우재혁,현성열,임용수 대한응급의학회 2013 大韓應急醫學會誌 Vol.24 No.5
Purpose: Studies have demonstrated that B-type natriuretic peptide (BNP) has a predictive value for sudden cardiac arrest in heart failure patients. The aim of this study was to investigate the usefulness of serum BNP levels for theprediction of death in post-cardiac arrest patients treated withtherapeutic hypothermia (TH). Methods: Out-of-hospital cardiac arrest (OHCA) survivors treated with TH between April 2007 and August 2010 were evaluated and divided into two groups based on death within 3 months. Initial serum BNP levels were checked and BNP levels compared between both groups. Results: A total of 162 patients were enrolled in the study. Among 162 patients, 109 patients were male (mean age of 50 years). The BNP levels of the non-survival group (n=77)were higher than those of the survival group (n=85); however,there was no statistical difference (19.45 pg/ml vs. 30.75pg/ml, p=0.174). The BNP cutoff value of 106 pg/ml for death within 3 months had a sensitivity of 35.1% and a specificity of 78.8%. In the logistic regression analysis, BNP levels higher than 106 pg/ml were significantly associated with death within 3 months (odds ratio [OR], 2.625; 95%confidence interval [CI], 1.066-6.463) and other independent factors were BLS to ROSC (OR, 1.105; CI, 0.082-1.038), non-VF/VT (OR, 3.698; CI, 1.632-8.380), and APACHE II score (OR, 1.117; CI, 1.035-1.204). Conclusion: Initial Serum BNP levels are related with death within 3 months for patients that received TH after OHCA. However, additional randomized and controlled studies are needed.
[발표논문] Support Vector Machine을 이용한 고객구매예측모형
안현철,김경재,한인구 한국지능정보시스템학회 2004 한국지능정보시스템학회 학술대회논문집 Vol.- No.-
이른바 ’대중 맞춤(mass customization)’이 기업 경쟁력 강화에 있어 중요한 이슈로 대두됨에 따라 인터넷 기반의 e-CRM 에 대한 세간의 관심이 크게 높아지고 있다. 특히 e-CRM 의 여러 분야 중에서도 특정 고객이 어떤 상품 혹은 상품군을 구매할 것인지를 예측, 분류하는 ’고객구매예측모형의 구축’ 분야는 최근 많은 연구가 이루어지고 있으며, 이미 국내외 실무현장에서 널리 작용되고 있다. 이러한 ‘고객구매예측모형’에는 전통적인 데이터마이닝 기법이 활용되어 왔다. 하지만 전통적인 기법의 경우, 정확도가 상대적으로 떨어지거나 혹은 모형의 구축 및 유지관리가 어렵다는 문제가 총총 제기되어 왔다. 이에 본 연구에서는 기존 모형의 문제점을 개선하기 위한 대안으로, 인공신경망 수준 이상의 높은 예측력을 나타내면서 동시에 과대적합 문제를 완화할 수 있는 것으로 알려진 Support Vector Machine(SVM)을 재안하고자 한다. 본 연구에서는 고객구매예측모형으로서 SVM의 적합성을 판단하기 위하여 로지스틱 희귀분석, 인공신경망과 그 성과를 비교하였다. 그 결과, SVM이 다른 기생들에 비해 상대적으로 우수한 성과를 냐타냄을 확인할 수 있었다.