• The Korean Journal of Applied Statistics
• /
• v.12 no.2
• /
• pp.517-525
• /
• 1999

본 논문에서는 정준판별분석에서 분석자의 사전지식을 고려한 조건부 계량화(partial quantification)를 제안하고 있다. 연관된 수식을 라그랑지 승수법으로 유도하였으며 SAS/IML을 이용한 프로그램과 예제를 간략히 설명하였다.

• 한국데이터정보과학회:학술대회논문집
• /
• 2006.11a
• /
• pp.113-117
• /
• 2006

우리는 다양한 자료에서 유의미한 정보를 파악하기 위한 방법으로 다변량 분석 방법 중에서 정준판별분석, 로지스틱, 다층퍼셉트론 그리고 의사결정나무를 사용자 편의를 극대화하고 사용이 간단한 비주얼 베이직 6.0을 이용하여 개발하였다.

• Bulletin of Food Technology
• /
• v.10 no.1
• /
• pp.94-101
• /
• 1997

NIR(근적외) 분광분석법이 녹차의 원산국을 판별하는데 이용할 수 있는지를 알아보기 위하여 분쇄한 47종의 한국산 및 일본산 녹차에 대하여 NIR 분석을 실시한 후, 그 분광 데이터에 대하여 principal component analysis(주성분 분석 )와 canonical variate analysis(정준판별분석)을 실시하였다. 15개의 주성분과 1100~2500nm에서의 first derivative log(1/R) 데이터를 이용할 경우, 제1 및 제2 정준판별함수는 한국산 녹차 및 일본산 녹차를 판별하는데 가장 효과적이었다. 사용된 canonical variate analysis는 녹차 시료를 97.87%의 정확도로 그 지리적 출처를 판별하였다. 한편 first derivative log(1/R) spectra상의 파장범위 1674~1686, 1950~1992, 2014~2030및 2118~2158 nm에서 일본산 녹차와 3종의 한국산 녹차 그룹간에 현저한 차이가 발견되었다. 이 차이는 polyphenols, caffeine 및 amino acids와 같은 녹차의 주요성분과 관련되어 있지 않으며 주로 지리적 출처상의 차이에 기인한 것으로 판단되었다.

• Bulletin of Food Technology
• /
• v.9 no.4
• /
• pp.87-93
• /
• 1996

NIR(근적외) 분광분석법이 참기름의 원산국 판별에 이용 가능 한지를 알아보기 위하여 32종의 시료에 대하여 NIR 분석을 실시한 후, 그 분광 데이터에 대하여 principal component analysis(주성분 분석)와 canonical variate analysis(정준판별분석)을 실시하였다. 10개의 주성분과 400-2500nm에서 second derivative log(1/R) 데이터를 이용할 경우, 제1 및 제2 정준판별함수는 3개 참기름 그룹(한국산 참깨로 제조한 13종의 참기름 그룹, 외국산 참깨로 제조한 10종의 국산 참기름 그룹 및 미지의 참깨로 제조한 9종의 참기름 그룹)을 판별하는데 가장 효과적이었다. 사용된 canonical variate analysis는 참기름 시료를 100%의 정확도로 그 지리적 출처를 분류하였다. 한편 second derivative log(1/R) spectra상의 파장범위 498-500, 668, 1698-1724, 2242-2256, 2302-2306, 2328 및 2348~2352nm에서 3개 그룹간에 현저한 차이가 발견되었다.

• Food Science of Animal Resources
• /
• v.29 no.1
• /
• pp.132-139
• /
• 2009

The objective of this study was to investigate the most effective analysis methods for palatability grading of Hanwoo beef by comparing the results of discriminant analysis with sensory data. The sensory data were obtained from sensory testing by 1,300 consumers evaluated tenderness, juiciness, flavor-likeness and overall acceptability of Hanwoo beef samples prepared by boiling, roasting and grilling cooking methods. For the discriminant analysis with one factor, overall acceptability, the linear discriminant functions and the non-parametric discriminant function with the Gaussian kernel were estimated. The linear discriminant functions were simple and easy to understand while the non-parametric discriminant functions were not explicit and had the problem of selection of kernel function and bandwidth. With the three palatability factors such as tenderness, juiciness and flavor-likeness, the canonical discriminant analysis was used and the ability of classification was calculated with the accurate classification rate and the error rate. The canonical discriminant analysis did not need the specific distributional assumptions and only used the principal component and canonical correlation. Also, it contained the function of 3 factors (tenderness, juiciness and flavor-likeness) and accurate classification rate was similar with the other discriminant methods. Therefore, the canonical discriminant analysis was the most proper method to analyze the palatability grading of Hanwoo beef.

• Korean Journal of Food Science and Technology
• /
• v.26 no.5
• /
• pp.561-566
• /
• 1994

The kinds of kochujang were discriminated by using their organic acid compositions, GC peak areas of volatile concentrates and flavor intensity determined by sensory evaluation. Tested kochujang were 51 kinds of traditional kochujang and 10 kinds of industry-produced kochujang in the market. The traditional kochujang included 20 kinds of Sunchang kochujang prepared with glutinous rice, 11 kinds of Boeun kochujang prepared with barley and 20 kinds of Sachun kochujang prepared with wheat. Boeun kochujang was distinguished from other kinds of traditional and industry-produced kochujang by using canonical discriminant analysis for the compositions of organic acids. Among organic acids, lactic acid was the most contributing variable for the discrimination of various kochujang. Traditional and industry-produced kochujang could be classified into 4 groups by using discriminant analysis for GC peak areas. The peak number 2, 4, 8 and 11 were found to be highly contributing variables for the discrimination of kochujang by using stepwise discriminant analysis. Industry-produced kochujang was discriminated from traditional kochujang by using canonical discriminant analysis for the intensity of 8 kinds of flavor property. The taste 'umami' was found to be the most contributing variable for the discrimination of kochujang.

• The Korean Journal of Applied Statistics
• /
• v.30 no.1
• /
• pp.83-93
• /
• 2017

A typical approach to visualizing k (${\geq}2$)-group multidimensional data is to use Fisher's canonical discriminant analysis (CDA). CDA finds the best low-dimensional subspace that accommodates k group centroids in the Mahalanobis space. This paper proposes an alternative visualization procedure functioning in the Euclidean space, which finds the primary dimension with maximum discrimination of k group centroids and the secondary dimension with maximum dispersion of all observational units. This hybrid procedure is especially useful when the number of groups k is two.

• Korean Journal of Food Science and Technology
• /
• v.40 no.2
• /
• pp.135-140
• /
• 2008

In this study, the macromineral content ratios of four herbal medicine samples(Saposhnikoviae Radix, Bupleuri Radix, Cnidii Rhizoma, and Astragali Radix) were analyzed to discriminate their geographical origins using an energydispersive x-ray fluorescence (EDXRF) technique. EDXRF is a rapid, non-destructive, and multi-elemental analysis technique. Initially, samples of both domestic and imported herbal medicines were pulverized, and then their macromineral contents, including P, S, K, and Ca, were analyzed using EDXRF. For the discrimination of their geographical origins, canonical discriminant analysis was carried out based on the estimated macromineral relative content ratios of the samples. According to the results, the discrimination accuracies were as follows: 93.3% for Saposhnikoviae Radix, 95.7% for Bupleuri Radix, 98.8% for Cnidii Rhizoma, and 87.5% for Astragali Radix. Overall, the results imply that this technique could be used as a standard method, to discriminate their geographical origins between domestic and imported herbal medicines.

• The Korean Journal of Applied Statistics
• /
• v.21 no.6
• /
• pp.969-980
• /
• 2008

This paper presents the comparison of four methods, linear, quadratic, canonical and non-parametric discriminant analyses to discriminate the consumers' taste grade with sensory variables, such as tenderness, juiciness, flavor, and overall acceptability based on Consumer Sensory Survey. The classification ability of each method is measured and compared by the resubstitution error rate.

• Korean Journal of Food Science and Technology
• /
• v.35 no.4
• /
• pp.653-659
• /
• 2003

A 20 MHz pulsed NMR systems was employed to discriminate the geographical origin of soybeans and black beans (yak-kong) from Korea and foreign countries. Crude fat contents measured by soxhlet method were significantly (p<0.05) different between domestic and imported soybeans. Moisture and crude protein contents, measured by AOAC, were significantly different between domestic and imported black beans. In soybeans, values by solid fat content method and Carr-Purcell-Meiboom-Gill (CPMG) method using 20 MHz pulsed NMR showed the significant difference among soybeans from various the geographical origins. In black beans (yak-kong), NMR values measured by NMR except $T_1$ SR pulse sequence revealed the significant difference by the geographical origins. The habitat of soybeans and black beans could be identified by canonical discriminant analysis of chemical composition with $70{\sim}91.7\;%$ accuracy. Low field NMR data followed by discriminant analysis, however, granted the 100% of accuracy for classification of soybeans.