• 에너지공학
• /
• 제25권4호
• /
• pp.190-197
• /
• 2016

인화점은 가연성 액체 용액을 안전하게 취급하기 위한 중요한 성질 중 하나이다. 본 논문에서는 Seta flash 밀폐식 장치를 이용하여 이성분계 용액인 water+1-propanol과 water+2-propanol계의 인화점을 측정하였다. 회귀 분석법을 이용하여 인화점을 계산하였다. 또한 라울의 법칙을 이용하여 인화점을 계산하였고, van Laar 식의 이성분계 파라미터를 최적화시키는 방법을 통해 인화점을 예측하였다. 각 인화점 계산 결과와 측정 결과를 비교하였다. 그 결과, 회귀 분석법에 의한 인화점 계산치가 가장 측정치를 잘 모사하였다.

• Journal of the Korean Data and Information Science Society
• /
• 제7권2호
• /
• pp.203-210
• /
• 1996

A linear neural unit with a modified anti-Hebbian learning rule has been shown to be able to optimally fit curves, surfaces, and hypersurfaces by adaptively extracting the minor component of the input data set. In this paper, we study how to use the robust version of this neural fitting method for linear regression analysis. Furthermore, we compare this method with other methods when data set is contaminated by outliers.

• 한국농공학회논문집
• /
• 제49권3호
• /
• pp.63-68
• /
• 2007

The LRA(Logistic Regression Analysis) conducts a quantitative analysis by collecting a lot of samples and the AHP(Analytic Hierarchy Program) makes use of expert decision influenced by subjective judgment to a certain degree. This study is to suggest a combination method in mapping landslide hazard by giving equal weight for the result of LRA and AHP. Topographic factors(slope, aspect, elevation), soil dram, soil depth and land use were adopted to classify landslide hazard areas. The three methods(LRA, AHP, the combined approach) was applied to a $520km^2$ region located in the middle of South Korea which have occurred 39 landslides during 1999 and 2003. The suggested method showed 58.9% matching rate for the real landslide sites comparing with the classified areas of high-risk landslide While LRA and AHP Showed 46.1% and 48.7% matching rates respectively. Further studies are recommended to find the optimal combining weight of LRA and AHP with more landslide data.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2003년도 추계학술대회
• /
• pp.1430-1435
• /
• 2003

Satellite upper platform is optimized by response surface method which has non-gradient, semi-glogal, discrete and fast convergency characteristics. Sampling points are extracted by design of experiments using Central Composite Method and Factorial Design. Also response surface is generated by the various regression functions. Structure analysis is execuated with regard for static and dynamic environment in launching stage. As a result response surface method is superior to other optimization method with respect to optimum value and cost of computation time. Also a confidence is varified in the various regression models.

• Journal of Computing Science and Engineering
• /
• 제11권4호
• /
• pp.121-129
• /
• 2017

Feature selection has been widely established as an efficient technique for microarray data analysis. Feature selection aims to search for the most important feature/gene subset of a given dataset according to its relevance to the current target. Unsupervised feature selection is considered to be challenging due to the lack of label information. In this paper, we propose a novel method for unsupervised feature selection, which incorporates embedded learning and $l_{2,1}-norm$ sparse regression into a framework to select genes in microarray data analysis. Local tangent space alignment is applied during embedded learning to preserve the local data structure. The $l_{2,1}-norm$ sparse regression acts as a constraint to aid in learning the gene weights correlatively, by which the proposed method optimizes for selecting the informative genes which better capture the interesting natural classes of samples. We provide an effective algorithm to solve the optimization problem in our method. Finally, to validate the efficacy of the proposed method, we evaluate the proposed method on real microarray gene expression datasets. The experimental results demonstrate that the proposed method obtains quite promising performance.

• 디지털콘텐츠학회 논문지
• /
• 제19권7호
• /
• pp.1357-1363
• /
• 2018

정보 기술의 발전과 더불어 인공 지능 및 기계 학습 분야는 다양한 응용 분야에서 성능을 인정받고 있으며, 다양한 응용 분야로 확대되고 있다. 본 논문에서는 기계 학습 방법을 응용한 소프트웨어 분석 방법을 제안한다. 소프트웨어의 특성을 표현하기 위해 소프트웨어의 코드 분포를 분석하고 이 정보를 기계 학습 방법인 선형 회귀를 통해 분석함으로써 유사 소프트웨어를 분석할 수 있는 방법을 제안한다. 소프트웨어의 특성은 프로그램 내에 포함된 명령어에 의해 표현될 수 있으며, 명령어의 분포 정보를 학습 데이터로 활용하였다. 또한, 학습 데이터를 통한 학습 과정은 소프트웨어 유사성 분석을 위한 선형 회귀 모델을 구성한다. 본 논문에서 제안한 방법은 구현 및 실험을 통해 정확성을 검증한다. 본 논문에서 제안한 방법은 소프트웨어의 유사성을 판단할 수 있는 기본 기술로 활용될 수 있을 것으로 기대된다. 또한 기계 학습 방법을 통한 소프트웨어 분석 기술에 응용될 수 있을 것으로 기대된다.

• Communications for Statistical Applications and Methods
• /
• 제16권2호
• /
• pp.317-323
• /
• 2009

Recently not even government polices but small town decisions are based on the survey data/information, so the most of government agencies/organizations demand various sample surveys in each fields for more detail information. However in conducting the sample survey, nonresponse problem rises very often and it becomes a major issue on judging the accuracy of survey. For that matters, one solution ran be using the administration data. However unfortunately most of administration data are restricted to the common users. The other solution can be the imputation. Therefore several method, of imputation are studied in various fields. In this study, in stead of the simple regression imputation method which is commonly used, factor score regression method is applied specially to the incomplete data which have the unit and item misting values in survey data. Here for simulation study, Consumer Expenditure Surveys in Korea are used.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2006년도 춘계학술대회 논문집
• /
• pp.202-205
• /
• 2006

The objective of this study is to analyze the hazard-areas for landslide using GIS and RS. LRA (Logistic Regression Analysis) and AHP (Analytic Hierarchy Program) methods were used for evaluation of the hazard-areas by six topographic factors (slope, aspect, elevation, soil drain, soil depth, land use). These methods were applied to Anseong-si where frequent landslides were occurred mainly by the regional heavy rainfall. A landslide hazard-map of Anseong-si could describe into 7 hazard-grades. As results, LRA method was underestimated in higher grades areas, while AHP method was underestimated in lower grades areas. In order to evaluate the hazard-areas for landslides with accuracy, these results of each method were overlapped and the results of suggested method were compared with the historical landslide hazard records of KFRI (Korea Forest Research Institute).

• 한국운동역학회지
• /
• 제31권2호
• /
• pp.148-153
• /
• 2021

Objective: It is to extract training items that have a high relationship with the improvement of weightlifting records through correlation and regression analysis between training methods used commonly in the field and Snatch records and jerk records. Through this, it is intended to promote training efficiency to improve the records of weightlifters. Method: For 90 elite weightlifters of the professional teams, 4 groups (lightweight (30 people): 61 kg, 67 kg, 73 kg., middleweight (30 people): 81 kg, 89 kg, 96 kg., heavyweight (30 people): 102 kg, 109 kg, +109 kg., the whole group (90 people)) were divided. At the significance level of 0.05, correlation analysis and linear regression analysis were performed between record of training methods used widely in the field and Snatch records and Jerk records. Results: First, the better the record in Jerk, the better the Snatch record. Second, the three training methods HS, ForceS and WP performed in the field were all found to be important factors related to the improvement of Snatch record. Third, In the jerk where there are more types of training than Snatch, three training methods (HC, ForceS, BPP) appeared to be an important training method for improving the jerk record. Conclusion: While many training methods have been devised and carried out in the field, 3 types of training (HS, ForceS, WP) for improving Snatch record and 3 types of training (HC, ForceS, BPP) for improving Jerk record was found to be the most influential training method. Since all of them showed a large value of explanatory power by regression analysis, it is considered that this study is meaningful in that it can promote training efficiency by simplifying although there are many types of training for athletes.

• 복식
• /
• 제54권3호
• /
• pp.73-85
• /
• 2004

The purpose of this study was to investigate the characteristic and sensibility of Japanism fashion designs which represented by Japanese designers and Western designers. The stimulus were 29 pictures of contemporary fashion designs which represented the Japanism style fashion designs from fashion collections. The data were analyzed by Cluster analysis, Factor analysis, Multidimensional Scaling Method and Regression Analysis. The specific objectives were as follows ; 1) As result of design analysis, Japanism fashion sensibility is unique and good-looking. 2) As result of the factor analysis. 4 factors which are Attractiveness, Attention, Maturity and Hardness and softness. 3) According to sensibility positioning, The Japanism fashion design was classified by Decorative-Simple, Hard-Soft. 4) As result of the Regression Analysis, The preference of Japanism fashion design was related to attractive factor. 5) As result of the Regression Analysis. The buying desirable of Japanism fashion design was related to attractive, attentive and mature factor.