• 약학회지
• /
• 제42권6호
• /
• pp.545-551
• /
• 1998

Near Infrared transmittance Spectroscopy (NIRS) was used to evaluate and quantify the pharmaceutical active compounds. In the paper, taurine (2-Aminoethanesulfonic acid) was quantitatively analyzed in commercial pharmaceutical preparations. For calibration a central composite factorial design was used to determine concentrations of ingredients in reference samples. For the quantitative analysis of taurine, the most suitable data analysis method includes the calculation of second derivatives and a partial least squares regression (PLSR) model. By NIR spectrometry, combined with PLSR, the taurine concentration was successfully predicted with a relative standard error of prediction (SEP) lower than 1.04%.

• Environmental Analysis Health and Toxicology
• /
• 제15권1_2호
• /
• pp.39-43
• /
• 2000

Linear alkyl sulfonates (LAS) constitute a large fraction of the surfactants used in cleaning processes in households, trade and industry Despite the industrial significance and the possible environmental impact of these compounds, the fast and inexpensive determination of LAS concentrations is still a difficult task. In this study, near infrared (NIR) spectroscopy which is a rapid spectroscopic analysis method compared with a traditional analytical method for the measurement of LAS concentration such as HPLC, GC and standard wet chemistry method. NIR spectra of LAS between 0.313 and 25.0% (w/v) in water were utilized to develop a calibration model. The best results (R ＝ 0.998, SEP ＝ 0.244% (w/v)) obtained by using partial least-squares regression with spectral data treatment and 2nd derivatization were comparable to the results (SEC ＝ 0.186% (w/v), SEP ＝ 0.206% (w/v)) obtained by using multiple linear least-squares regression (MLR). However, models based on derivative spectra have no significant advantage with MLR.

• 대한인간공학회지
• /
• 제19권3호
• /
• pp.23-37
• /
• 2000

Usability is one of the most important factors that affect customers' decision to purchase a product. Several studies have been conducted to model the relationship between the product design variables and the product usability. Since there could be hundreds of design variables to be considered in the model, a variable screening method is required. Traditional variable screening methods are based on expert opinions (Expert screening) in most Kansei engineering studies. Suggested in this study are statistical methods for screening important design variables by using the principal component regression(PCR), cluster analysis, and partial least squares(PLS) method. Product variables with high effect (PCR screening and PLS screening) or representative variables (Cluster screening) can be used to model the usability. Proposed variable screening methods are used to model the usability for 36 audio/visual products. The three analysis methods (PCR, Cluster, and PLS) show better model performance than the Expert screening in terms of $R^2$, the number of variables in the model, and PRESS. It is expected that these methods can be used for screening the product design variables efficiently.

• Asian Pacific Journal of Cancer Prevention
• /
• 제14권11호
• /
• pp.6347-6350
• /
• 2013

Post-transplant lymphoproliferative disorder (PTLD) is a common complication of therapeutic immunosuppression after organ transplantation. Gene expression profile facilitates the identification of biological difference between Epstein-Barr virus (EBV) positive and negative PTLDs. Previous studies mainly implemented variance/regression analysis without considering unaccounted array specific factors. The aim of this study is to investigate the gene expression difference between EBV positive and negative PTLDs through partial least squares (PLS) based analysis. With a microarray data set from the Gene Expression Omnibus database, we performed PLS based analysis. We acquired 1188 differentially expressed genes. Pathway and Gene Ontology enrichment analysis identified significantly over-representation of dysregulated genes in immune response and cancer related biological processes. Network analysis identified three hub genes with degrees higher than 15, including CREBBP, ATXN1, and PML. Proteins encoded by CREBBP and PML have been reported to be interact with EBV before. Our findings shed light on expression distinction of EBV positive and negative PTLDs with the hope to offer theoretical support for future therapeutic study.

• 응용통계연구
• /
• 제31권2호
• /
• pp.253-263
• /
• 2018

본 연구에서는 우리나라 청소년의 인터넷 중독 자료를 희소 부분 최소제곱법을 적용 하여 분석하였다. 서울 성모병원에서 수집된 자가보고 방식의 인터넷 중독 측도와 다양한 임상 및 정신 병리학적 설문 문항들을 자료로 활용하였다. 표본의 개수보다 설문 문항의 수가 많은 고차원 자료이며, 각 세부문항끼리는 상관관계가 높아 부분 최소제곱법이 좋은 회귀분석 모형이다. 보다 높은 예측 성능과 해석력을 얻기위해 희소성 제약 조건이 가능한 희소 부분 최소 제곱법을 이용하였고, 2가지 다른 벌칙함수를 이용하여 가장 좋은 방법을 선택하였다. 또한, 제안된 최종 모형을 통해 인터넷 중독이 임상 및 정신 병리학적 측도들로 잘 설명됨을 보이고, 공격성과 관련된 다른 설문 문항이 설문 문항이 모형의 잠재성분을 구별하고 설명하는데 역할을 한다는 유의미한 결과를 도출하였다.

• 분석과학
• /
• 제15권3호
• /
• pp.243-247
• /
• 2002

인근적외선 분광법(Near infrared spectroscopy, NIRS)으로 얻은 디아제팜정의 스펙트럼을 부분최소자승법(Partial least squares regression, PLSR)으로 처리하여 디아제팜정에서 디아제팜을 신속하고 간단하게 분석하는 방법을 연구하였다. 2 mg 및 5 mg 의 디아제팜정을 이용하여 검정선을 작성하고, 이 검정선의 직선성, 정량범위, 재현성 등을 검토하여 본 분석법의 정확성을 검증하였다. 2 mg의 디아제팜의 검정선의 상관계수는 0.9416이고, 검정선 표준오차(SEC)는 0.018%이였다. 5 mg의 디아제팜의 검정선의 상관계수는 0.9157이고, 검정선오차는 0.032%이였다.

• 응용통계연구
• /
• 제10권2호
• /
• pp.227-239
• /
• 1997

분광학분야에서 측정되는 자료는 여러 파장에서 측정된 스펙트럼 행렬과 이 스펙트럼을 통하여 알고자하는 어떤 반응치들의 행렬 또는 벡터로 주어진다. 이 경우 측정 자료에의 많은 잡음(noise)과 파장간의 상관관계가 내재한다. 부분최소제곱회귀 방법은 여러 개의 파장에서 측정된 자료를 모두 이용하는데 자료축약과정을 통하여 자료의 잡음 문제와 상관관계 문제를 해결하는 다변량통계방법이다. 본 연구에서는 이러한 자료에 적합한 부분최소제곱회귀의 이론을 알아보고 실제로 측정된 자료를 통하여 주어진 스펙트럼에 대한 반응치의 예측을 부분최소제곱회귀 방법을 이용하여 고찰하였다.

• 한국근적외분광분석학회:학술대회논문집
• /
• 한국근적외분광분석학회 2001년도 NIR-2001
• /
• pp.1276-1276
• /
• 2001

Polycyclic aromatic hydrocarbons(PAHs) are widely distributed in the environment and are often implicated as potential carcinogens. The chromatographic methods of detection and quantitative determination of PAHs in environmental samples are costly, time consuming, and do not account for all kinds of PAHs. This work describes a quantitative spectroscopic method for the analysis of mixtures of eight PAHs using multivariate calibration models for Fourier transform near infrared(FT-NIR) spectral data. The NIR spectra of mixtures of PAHs (anthracene, pyrene, 1,2-benzanthracene, perylene, chrysene, benzo(a)pyrene, 1-methylanthracene and benzo(ghi)perylene) were measured in the wavelength range from 1100 nm to 2500 nm. The spectral data were processed using a partial least squares regression. We have studied the spectral characteristics of NIR spectra of mixtures of PAHs. It was possible to determine each PAM used in this study at the environmental level(mg L-1) in the laboratory samples. Further development may lead to the rapid determination of more PAHs in typical environmental samples.

• 센서학회지
• /
• 제17권2호
• /
• pp.120-126
• /
• 2008

Glucose concentrations in the interstitial fluid were measured based on optical spectroscopy. Prediction of glucose concentrations was made using partial least squares regression and accuracy improvement was achieved by data preprocessing as well as by selecting an optimal wavelength region. For this purpose, artificial interstitial fluid samples were prepared where their glucose levels varied between 0 and 10 g/dl. Infrared spectral regions where glucose absorption lies were investigated. A region of 1000 - 1500 $cm^{-1}$ produced the best accuracy among the regions of 1000 - 1500 $cm^{-1}$, 4000 - 4545 $cm^{-1}$1 and 5500 - 6500 $cm^{-1}$. Further accuracy improvement in 1000 - 1500 $cm^{-1}$ was achieved by selecting specific wavelength bands based on a loading vector analysis method. For the samples whose glucose concentrations ranged between 0 and 0.5 g/dl, SEP= 0.0266 g/dl and R =0.9863 were achieved with 1000 - 1500 $cm^{-1}$. However, the loading vector optimized band of 1002 - 1095 $cm^{-1}$ reduced the prediction error up to 47 % (SEP =0.0125 g/dl and R=0.9970).

• 한국식품영양학회지
• /
• 제29권5호
• /
• pp.605-609
• /
• 2016

This study aimed to find the optimum instrumental test conditions for the Texture Profile Analysis (TPA) of cooked rice in order to predict the sensory texture attributes (hardness, adhesiveness, chewiness). Sensory evaluation was performed for three kinds of instant cooked rice with university students in their twenties and the results of the sensory evaluation were compared to instrumental TPA patterns. Using partial least squares regression, the instrumental TPA results at a cross-head speed of 1.0 mm/sec and a compression ratio of 70% proved to be an excellent predictor of the sensory attributes of hardness ($R^2=0.99$) and chewiness ($R^2=0.99$). The results at a cross-head speed of 0.5 mm/sec and compression ratio of 30% provided an excellent model for the prediction of sensory adhesiveness ($R^2=0.83$). In this experimental range, sensory hardness and chewiness showed a high correlation with instrumental TPA parameters (hardness, cohesiveness, adhesiveness, springiness, chewiness) with a high cross-head speed and compression ratio, while sensory adhesiveness showed a high correlation with the TPA parameters with a low cross-head speed and compression ratio.