• 응용통계연구
• /
• 제34권3호
• /
• pp.401-410
• /
• 2021

판별분류분석에서 널리 이용되는 k-최근접 이웃 분류 방법은 고정된 이웃의 수만을 고려하여 자료의 국소적 특징을 반영하지 못하는 한계가 있다. 이에 자료의 국소적 구조를 고려하여 이웃의 개수를 선택하는 적응 최근접이웃방법이 개발된 바 있다. 고차원 자료의 분석에 있어서는 k-최근접 이웃 분류를 사용하기 전에 랜덤 투영 기법 등을 활용하여 차원 축소를 수행하는 것이 일반적이다. 이렇게 랜덤 투영시킨 다수의 분류 결과들을 면밀히 조합하여 투표를 통해 최종 할당을 하는 기법이 최근 개발된 바 있다. 본 연구에서는 고차원 자료에서의 분석을 위해 적응 최근접이웃방법과 랜덤 투영 앙상블 기법을 조합한 새로운 판별분류 기법을 제안하였다. 제안된 방법은 기존에 개발된 방법에 비해 분류 정확성 측면에서 더 뛰어남을 모의실험 및 실제 사례 분석을 통해 확인하였다.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 1996년도 춘계학술대회 논문집
• /
• pp.47-51
• /
• 1996

Unsteady-state temperature distributions and thermal deformations of a spindle system are studied in this paper. Three dimensional model is built for analysis, and the amount of heat generation of bearing and the thermal characteristic values including heat transfer coefficient are estimated. Temperature distributions and thermal deformations of a model are analyzed using the finite element method and the termal boundary values. Numerical results are compared with the measured data. The results show that thermal deformations and temperature distributions of a high precision spindle system can be reasonably estimated using the three dimensional model and the finite element method.

• Communications for Statistical Applications and Methods
• /
• 제3권1호
• /
• pp.11-19
• /
• 1996

Canonical correlation analysis is a multivariate technique for identifying and quantifying the statistical relationship between two sets of variables. Like most multivariate techniques, the main objective of canonical correlation analysis is to reduce the dimensionality of the dataset. It would be particularly useful if high dimensional data can be represented in a low dimensional space. In this study, we will construct statistical graphs for paired sets of multivariate data. Specifically, plots of the observations as well as the variables are proposed. We discuss the geometric interpretation and goodness-of-fit of the proposed plots. We also provide a numerical example.

• 패션비즈니스
• /
• 제15권6호
• /
• pp.101-112
• /
• 2011

Three-dimensional(3D) virtual clothing simulation system may require the use of physical, mechanical, and configurational data in order to mimic the actual clothing with high degree of realism. Therefore the 3-dimensional scanning system based on optical methods was adopted to extract the 3-dimensional data of the fabric surface. In this study, the appearances of the 3-dimensionally transformed textile fabrics via several finishing procedures were investigated using a 3D scanning system. The wool gauze fabrics treated with the shrink-proof finishing and the felting process showed height changes up to 4.5mm. The 3-dimensional configuration may be objectively described by the use of mesh generation from the scanned output. The generated mesh information may further be utilized in the 3D virtual clothing simulation system for accurate description of the fashionable textile materials used in the simulation system.

• 한국가시화정보학회지
• /
• 제21권1호
• /
• pp.18-25
• /
• 2023

In these days, high dimensional data prediction technology based on neural network shows compelling results in many different kind of field including engineering. Especially, a lot of variants of convolution neural network are widely utilized to develop pixel level prediction model for high dimensional data such as picture, or physical field value from the sensors. In this study, velocity vector field of ideal flow on ship surface is estimated on pixel level by Unet. First, potential flow analysis was conducted for the set of hull form data which are generated by hull form transformation method. Thereafter, four different neural network with a U-shape structure were conFig.d to train velocity vectors at the node position of pre-processed hull form data. As a result, for the test hull forms, it was confirmed that the network with short skip-connection gives the most accurate prediction results of streamlines and velocity magnitude. And the results also have a good agreement with potential flow analysis results. However, in some cases which don't have nothing in common with training data in terms of speed or shape, the network has relatively high error at the region of large curvature.

• 한국의류산업학회지
• /
• 제22권5호
• /
• pp.595-606
• /
• 2020

This study develops an optimal two-dimensional (2D) pattern from three-dimensional human scan data by considering the cycling posture and dermatome of high school male cyclists. By analyzing the body surface change in the cycling posture and considering the dermatome of the lower limbs, the optimal cutting line setting and the development of cycling tights for individual cyclists were presented to provide data that could be used in the clothing industry. We designed three cycling tights to solve the size unsuitability. 3D design 1 is a non-extension design based on the analysis of the 3D human body scan data, in which parts were connected diagonally from the front of the knee to the back of the knee. 3D design 2 removed both the front and back to reduce air resistance during cycling. 3D design 3 did not have a cutting line on the front panel because of the air resistance during cycling in the front area. We analyzed the garment pressure for 8 points of lower body and performed a subjective evaluation of the 3D designed tights and the current cycling tights. The 3D design 1 in this study was well received in the omphalion, thigh, and hip area, while 3D design 3 was well received in the omphalion, thigh, hip, and bottom bands. Therefore, the LoNE of 3D design 1 was applied to the front, and the hip cutting line of 3D design 3 was applied to the back.

• 디지털융복합연구
• /
• 제12권8호
• /
• pp.295-302
• /
• 2014

3차원 공간정보를 구축하기 위해서는 고정밀의 3차원 점군데이터를 취득할 수 있는 레이저스캐닝 기술과 고해상도의 다중분광 영상정보를 취득할 수 있는 사진측량용 카메라의 활용은 필수이다. 그러나 사진측량용 카메라는 장비특성상 높은 구입비와 어려운 구입경로, 낮은 적용성으로 폭넓은 활용분야에 비해 활용성이 떨어진다. 따라서 일반사용자가 빠르고 간편하게 접근할 수 있는 디지털카메라를 이용하여 다차원 정사영상을 구축하는 기법을 연구하였다. 즉 3차원공간정보의 핵심자료인 3차원 다중분광영상정보를 구축하기 위해 디지털카메라를 개조하고 캘리브레이션 작업을 수행하였다. 스테레오 사진측량을 위한 기준점 측량과 관측대상에 대한 다중분광촬영, 정사영상으로의 변환 등을 거쳐 다차원 정사영상을 구축하였다.

• 디지털융복합연구
• /
• 제15권1호
• /
• pp.247-253
• /
• 2017

피처선택, 혹은 변수 선택은 피처의 수가 매우 많은 고차원 데이터에서 주어진 주제와 연관성이 높은 피처를 선별하는 과정으로서, 데이터의 차원수를 낮추어 군집분석이나 분류 분석 등을 용이하게 하는데 중요한 기법이다. 많은 수의 피처들 중에서 일부의 피처를 선별하기 위해서는 피처들을 평가하기 위한 도구가 필요하다. 현재까지 제안된 도구들은 대부분 확률이론이나 정보이론에 기초하여 만들어졌기 때문에 하나의 피처, 즉 1차원 데이터만을 평가할 수 있다. 그러나 피처들 간에는 상호작용이 있기 때문에 하나의 피처를 평가하기 보다는 여러 피처들의 집합, 즉 다차원 데이터를 평가할 수 있어야 효과적인 피처 선택이 가능하다. 본 연구에서는 확장된 거리 함수를 이용하여 1차원 데이터 평가용으로 제안된 FSDD 평가 함수를 다차원 데이터에 대한 평가가 가능하도록 개선하는 방법에 대해 제안하였다. 본 연구에서 제안한 접근법은 다른 1차원 데이터 평가함수에도 적용이 될 수 있을 것으로 기대된다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.877-878
• /
• 2008

We suggest a feature reduction method to classify mouse function data sets, which integrate several biological data sets represented as high dimensional vectors. To increase classification accuracy and decrease computational overhead, it is important to reduce the dimension of features. To do this, we employed Hybrid Huberized Support Vector Machine with kernels used for a kernel logistic regression method. When compared to support vector machine, this a pproach shows the better accuracy with useful features for each mouse function.

• Journal of the Korean Data and Information Science Society
• /
• 제26권2호
• /
• pp.505-516
• /
• 2015

We consider sparse high-dimensional linear regression models. Penalized regressions have been used as effective methods for variable selection and estimation in high-dimensional models. In penalized regressions, it is common practice to standardize variables before fitting a penalized model and then fit a penalized model with standardized variables. Finally, the estimated coefficients from a penalized model are recovered to the scale on original variables. However, these procedures produce a slightly different solution compared to the corresponding original penalized problem. In this paper, we investigate issues on the standardization of variables in penalized regressions and formulate the definition of the standardized penalized estimator. In addition, we compare the original penalized estimator with the standardized penalized estimator through simulation studies and real data analysis.