http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
고창완,정영선,김재희 대한산업공학회 2020 대한산업공학회지 Vol.46 No.4
In clustering problems, it is difficult to know the optimum number of clusters for a given dataset a priori. Hence, cluster validity indices (CVIs) measuring the fitness of partitions produced by clustering algorithms are important criteria to evaluate the goodness of clustering results. However, many CVIs suffer from asymmetric, arbitrary, noise, and sub-cluster shape of clusters, especially for high-dimensional dataset. Therefore, this paper proposes new CVIs in feature space in which the proposed CVIs transform arbitrary shape of clusters into elliptical or circular clusters by using kernel functions. The experimental results show that the proposed CVIs have a good performance to estimate the optimum number of clusters for asymmetric, noise, and arbitrary shapes of clusters.
고창완,김현민,정영선,김재희 한국ITS학회 2020 한국ITS학회논문지 Vol.19 No.4
자동차는 우리의 일상에 필수재가 된 지 오래지만 자동차 교통사고로 인한 사회적 비용이 국가 예산의 9%를 넘을 정도로 심각하여 이에 대한 국가적인 예방 및 대응 체계 구축이 매우 필요한 실정이다. 이에 본 연구에서는 빅데이터 분석 기법을 활용하여 차대차 교통사고의 상 해 심각도를 정확히 예측할 수 있는 모형을 제시하고자 하였다. 이를 위해 과거 3년간의 전국 교통사고 발생 데이터를 토대로, K-최근접 이웃, 로지스틱 회귀분석, 나이브베이즈, 의사결정 나무, 앙상블 알고리즘을 적용하여 각 모델의 상해 심각도 분류의 성능을 비교 분석하였다. 특 히 이 과정에서 각 상해 심각도 수준 간의 데이터 수에 차이가 있음에 주목하여 표본수가 많은 그룹에 대해서는 과소표본추출을 시행하는 등의 방법을 통해 분류 예측의 정확도를 높일 수 있었고, 분산 분석을 통해 모델의 유의성을 검증하였다. Automobiles have long been an essential part of daily life, but the social costs of car traffic accidents exceed 9% of the national budget of Korea. Hence, it is necessary to establish prevention and response system for car traffic accidents. In order to present a model that can classify and predict the degree of injury in car traffic accidents, we used big data analysis techniques of K-nearest neighbor, logistic regression analysis, naive bayes classifier, decision tree, and ensemble algorithm. The performances of the models were analyzed by using the data on the nationwide traffic accidents over the past three years. In particular, considering the difference in the number of data among the respective injury severity levels, we used down-sampling methods for the group with a large number of samples to enhance the accuracy of the classification of the models and then verified the statistical significance of the models using ANOVA.
고창완,이현종,박희석 제주대학교 농과대학 제주도축산문제연구소 1986 畜産論叢 Vol.1 No.1
This study was carried out to determined the physico-chemical properties of milk protein of Cheju native mare milk using polyacrylamide gel and sodium dodecyl sulfate -polyacrylamide gel elrctroporesis and to obtain fundamental information for the preservation of Cheju native horse. The results obtained are summarized as follows; 1. Whey protein of Cheju native mare's milk were separated on polyacrylamide gel electrophoresis into immunoglobulin, serum albumin, α-lactalbumin. β-lactoglobulin , whey protein. 1) The whey protein showed four variable bands which were considered to be different types (A,B, C and D). 2) Seven whey phenotypes (AA,BB,AB,AC,BD,BC and CD) were observed. The distribution of AA, AB and BB was 13.4% 40% and 13.4%, respectively, and that of the other individual phenotypes was less than 10%. 2. Casein of Cheju native mare's milk was separated into three malor components α_(s)-.β-and k-casein on polyacrylamide gel electrophoresis. 1) α_(s)-casein zone was found in as hort distance wlth 4-7 zones. 2) The ratio of α_(s)-casein to β-casein was 1:1.3 and simllar to that reported on thorughbred mare milk (1:1.6) but higher than that β-casein found in Holste in cow's milk (1:0.6) 3) k-casein was found the form of a polymer held together by intermolecular dlsulfide bond. 3. Whey protein of Cheju native mare's milk were separated on sodium dodecyl sulfate-polyacrylamide gel electrophoresis in to 7 bands. The molecular weights of correspond to approximately 66,500, 66,000, 65,000, 59,500, 19,500, 16,500, 12,000 daltons, respectively.
임성연,고창완,정영선 (사)한국스마트미디어학회 2022 스마트미디어저널 Vol.11 No.3
Researches on the prediction of domestic apartment sales price have been continuously conducted, but it is not easy to accurately predict apartment prices because various characteristics are compounded. Prior to predicting apartment sales price, the analysis of major factors, influencing on sale prices, is of paramount importance to improve the accuracy of sales price. Therefore, this study aims to analyze what are the factors that affect the apartment sales price in Gwangju, which is currently showing a steady increase rate. With 6 years of Gwangju apartment transaction price and various social factor data, several maching learning techniques such as multiple regression analysis, random forest, and deep artificial neural network algorithms are applied to identify major factors in each model. The performances of each model are compared with RMSE (Root Mean Squared Error), MAE (Mean Absolute Error) and (coefficient of determination). The experiment shows that several factors such as ‘contract year’, ‘applicable area’, ‘certificate of deposit’, ‘mortgage rate’, ‘leading index’, ‘producer price index’, ‘coincident composite index’ are analyzed as main factors, affecting the sales price. 국내 아파트 매매가 예측에 관한 연구는 현재까지 지속적으로 수행되어 왔지만, 아파트 가격은 다양한 특성이 복합적으로 작용하기 때문에 예측하는데 어려움을 겪고 있다. 아파트 매매가를 예측하는데 앞서 정확도를 높이기 위해서는 주요 변수 선정 및 영향요인 분석이 무엇보다 중요하다. 이에 본 연구는 현재 꾸준한 상승률을 보이는 광주광역시를 대상으로 아파트 매매가에 영향을 주는 요인을 분석해보고자 한다. 이를 위해 6년간의 광주광역시 아파트 실거래가와 각종 사회적 요인 데이터를 토대로, 다중회귀분석, 랜덤 포레스트, 심층인공신경망 알고리즘을 적용하여 각 모델에서 주요 영향요인을 파악하였으며, 모델의 성능은 평균 제곱근 오차, 평균 절대 오차 그리고 결정계수를 통해 비교 분석하였다. 본 연구에서는 딥러닝의 일종인 심층인공신경망의 성능이 가장 우수함을 보였고, 매매가에 영향을 미치는 주요 요인으로 건축경과연수, 계약연도, 적용면적, 양도성예금증서, 주택담보대출금리, 선행지수, 생산자물가지수, 동행지수 등이 도출되었다.
제주한우 교잡종의 혈청단백 유전적 변이 구명에 관한 연구
김동철,이희석,김중계,이현종,고창완,문성호,오운용 제주대학교 농과대학 제주도축산문제연구소 1988 畜産論叢 Vol.3 No.1
This paper identifies genetic distance between breeds in order to clarify genetic constitutions of livestock population using electrophoresis of polyacrylamide gels. The results of this study are surrmarised as follows ; The frequencies on the albumin aleles (AIb^(F), AI^(S)) were 0.800 and 0.200 in Cheju Native Cattle; 0.500 and 0.449 in BK in (Brahman X Cheju Native Cattle): 0.467 and 0.553 in B^(2)K;0.500 and 0.500 in cross breds of B^(3)K. The frequencies of heterozygote per locus of albumin in Cheju Native Cattle. Brahman X Cheju Native Cattle crossbred were 0.320 and 0.495~0.500 respectively. The frequencies of the transferrin alleles were Tf^(D) 0.600 and Tf^(E), 0.400 in Cheju Native Cattle. In crossbreds, the frequencies observed (Tf^(A), and Tf^(E)) were 0.095. 0.435 and 0.472 for the BK, 0.83 0.475 and 0.442 for the B^(2)K ; 0.042, 0.583, and 0.375 for the B^(3)K. The frequencies of the heterozygote per locus of transferrin were 0.048 in Cheju Native Cattlr College of Agriculture, Chelu National University) and 0.579, 0.572, 0.518 in crossbred of BK, B^(2)K and B^(3)K respectively.