• Journal of the Korea Safety Management & Science
• /
• v.15 no.4
• /
• pp.311-315
• /
• 2013

본 논문에서는 주성분 회귀법과 부분최소자승 회귀법을 비교하여 보여준다. 이 비교의 목적은 선형형태를 보유한 근적외선 분광 데이터의 분석에 사용할 수 있는 적합한 예측 방법을 찾기 위해서이다. 두 가지 데이터 마이닝 방법론인 주성분 회귀법과 부분최소자승 회귀법이 비교되어 질 것이다. 본 논문에서는 부분최소자승 회귀법은 주성분 회귀법과 비교했을 때 약간 나은 예측능력을 가진 결과를 보여준다. 주성분 회귀법에서 50개의 주성분이 모델을 생성하기 위해서 사용지만 부분최소자승 회귀법에서는 12개의 잠재요소가 사용되었다. 평균제곱오차가 예측능력을 측정하는 도구로 사용되었다. 본 논문의 근적외선 분광데이터 분석에 따르면 부분최소자승회귀법이 선형경향을 가진 데이터의 예측에 가장 적합한 모델로 판명되었다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.12A
• /
• pp.1781-1786
• /
• 2000

본 논문에서는, 적응 간섭 제거기(AIC : adaptive interference canceller)에 사용되는 적응 알고리즘 중 계산량이 적고, 하드웨어적 복잡성이 낮은 최소 평균 자승(LMS)알고리즘의 적응화 상수(constant step size)를 여러 개 사용하여 빠른 수렴 속도와 낮은 평균 자승 에러를 가지는 방법을 제안한다. 최소 평균 자승 알고리즘에서 적응화 상수는 수렴속도와 평균 자승 에러를 제거하는데, 적응화 상수가 증가할수록 수렴속도가 빨라지는 반면, 평균 자승 에러는 증가하게 된다. 이 논문에서는 수렴속도를 증가하는 동시에 평균 자승 에러를 줄이기 위해, 최소 평균 자승 알고리즘에서 세 개의 적응화 상수를 가지는 새로운 검출기를 제안한다. 이 구조에서, 매 반복횟수에 따른 각 그룹 출력 값들을 가지고, 선택(selection)부분에서 평균 자승 에러들을 비교하며, 가장 작은 평균 자승 에러를 나타내는 그룹의 에러 값과 필터 계수 값들이 선택되어져 여러 적응화 상수 최소 평균 자승 알고리즘(several step size LMS algorithm)부분에서 각 그룹의 필터 계수를 갱신하는데 필요한 정보로 이용된다.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.12
• /
• pp.3200-3210
• /
• 1997

In image processing we encounter some tasks to detect ellipse or to discriminate the curves. LSM is well used to fit curves to ellipse but it can fail to fit correctly when fitting to defected one. To overcome this problem, we propose Modified LSM. Only 2-parameters among 5-paramaters are to be determined by LSM, while 3-parameters are to be calculated by the constrain that the curve must pass 3 given points. Those 3 points are selected by operator so as to have elliptic feature. Such proposed MLSM shows better result than genunal LSM in case when the ellipse is severely defected. and is proved to be good method for determing the human dentition.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.5
• /
• pp.931-940
• /
• 2011

Classification is to generate a rule of classifying objects into several categories based on the learning sample. Good classification model should classify new objects with low misclassification error. Many types of classification methods have been developed including logistic regression, discriminant analysis and tree. This paper presents a new classification method using penalized partial least squares. Penalized partial least squares can make the model more robust and remedy multicollinearity problem. This paper compares the proposed method with logistic regression and PCA based discriminant analysis by some real and artificial data. It is concluded that the new method has better power as compared with other methods.

• Korean Chemical Engineering Research
• /
• v.53 no.2
• /
• pp.236-242
• /
• 2015

We present two data-driven modeling methods, partial least square (PLS) and artificial neural network (ANN), to predict the major operating and performance variables of a polymer electrolyte membrane (PEM) fuel cell stack. PLS and ANN models were constructed using the experimental data obtained from the testing of a 30 kW-class PEM fuel cell stack, and then were compared with each other in terms of their prediction and computational performances. To reduce the complexity of the models, we combined a variables importance on PLS projection (VIP) as a variable selection method into the modeling procedure in which the predictor variables are selected from a set of input operation variables. The modeling results showed that the ANN models outperformed the PLS models in predicting the average cell voltage and cathode outlet temperature of the fuel cell stack. However, the PLS models also offered satisfactory prediction performances although they can only capture linear correlations between the predictor and output variables. Depending on the degree of modeling accuracy and speed, both ANN and PLS models can be employed for performance predictions, offline and online optimizations, controls, and fault diagnoses in the field of PEM fuel cell designs and operations.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.33 no.10
• /
• pp.26-34
• /
• 2005

In this paper, a novel moving least squares interface welding method is proposed to carry out the coupled analysis of whole model composed of independently modeled finite element substructures with nodal mismatching interfaces. To verify the validity, and efficiency of the proposed interface welding method, various numerical examples are worked out including patch tests, convergence tests, and examples of coupled analyses of the structural systems with mismatching substructures. From the numerical tests, it is confirmed that one can efficiently carry out the coupled analysis of whole model composed of mismatching finite element substructures through the proposed method without any remeshing or any additional unknown.

• The Korean Journal of Applied Statistics
• /
• v.23 no.2
• /
• pp.295-304
• /
• 2010

Coating thickness is one of target variables in quality control process in steel industry. To predict coating thickness and to control quality of anti-fingerprint steel coils, ultraviolet-visible spectra are measured. We propose a variable-interval selection procedure based on the variable importance in projection in partial least square model. Using the proposed variable interval selection method, prediction performance gets better in the reduced model than the full model with full spectra absorbance. It is also shown that the first differencing as a data preprocessing technique does work well for the prediction of coating thickness.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.34 no.4
• /
• pp.283-289
• /
• 2014

This study was conducted to develop a non-destructive detection method for adulterated powder products using Raman spectroscopy and partial least squares regression(PLSR). Garlic and ginger powder, which are used as natural seasoning and in health supplement foods, were selected for this experiment. Samples were adulterated with corn starch in concentrations of 5-35%. PLSR models for adulterated garlic and ginger powders were developed and their performances evaluated using cross validation. The $R^2_c$ and SEC of an optimal PLSR model were 0.99 and 2.16 for the garlic powder samples, and 0.99 and 0.84 for the ginger samples, respectively. The variable importance in projection (VIP) score is a useful and simple tool for the evaluation of the importance of each variable in a PLSR model. After the VIP scores were taken pre-selection, the Raman spectrum data was reduced by one third. New PLSR models, based on a reduced number of wavelengths selected by the VIP scores technique, gave good predictions for the adulterated garlic and ginger powder samples.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.6
• /
• pp.619-623
• /
• 2008

A lane detection system is one of the major components of intelligent vehicle systems. Difficulties in lane detection mainly come from not only various weather conditions but also a variety of special environment. This paper describes a simple and stable method for the broken lane tracking in various environments. Probabilistic Hough Transform (PHT) and the Least-square method (LSM) are used to track and correct the lane orientation. For the efficiency of the proposed method, two regions of interest (ROIs) are placed in the lower part of each image, where lane marking areas usually appear with less intervention in our system view. By testing in both a set of static images and video sequences, the experiments showed that the proposed approach yielded robust and reliable results.

• 한국가스학회:학술대회논문집
• /
• 2000.09a
• /
• pp.180-185
• /
• 2000