• 제목/요약/키워드: partial least square

검색결과 512건 처리시간 0.03초

A Rapid Quantitative Assay of Intact Ambroxol Tablets by FT-NIR Spectroscopy

  • Kim, Do-Hyung;Ah, Woo-Young;Kim, Hyo-Jin
    • 대한약학회:학술대회논문집
    • /
    • 대한약학회 2003년도 Proceedings of the Convention of the Pharmaceutical Society of Korea Vol.2-2
    • /
    • pp.213.2-213.2
    • /
    • 2003
  • A simple analytical procedure using FT-NIR for the rapid determination of individual ingredients was evaluated. Direct measurements were made by reflection using a reflectance accessory, by transmittance using tablet accessory and turn table. FT-NIR spectral data were transformed to the first derivative. Partial Least Square Regression(PLSR) was applied to quantify near-infrared (NIR) spectra of 2 ingredients. These calibration models were cross-validated (leave-one-out approach). The prediction ability of the models was evaluated on ambroxol tablets and compared with the real values in manufacturing procedure. (omitted)

  • PDF

근적외선 분광광도법을 이용한 페노바르비탈정제의 정량법에 관한 연구 (Determination of Phenobarbital in Intact Phenobabital Tablets using NIRS)

  • 차기원;제금련;윤미옥;이수정;최현철;김호정;김효진
    • 분석과학
    • /
    • 제15권2호
    • /
    • pp.102-107
    • /
    • 2002
  • 근적외선 분광광도법 (near infrared spectroscopy, NIRS)를 이용하여 의약품의 신속 정확한 분석법을 개발하였다. 제조회사에 따라 약효성분이 같고 유사한 부형제를 갖고 있는 페노바르비탈정제에서 페노바르비탈을 액체크로마토그래피로 분석한 분석치와 투과방식의 NIRS 의 부분최소자승법(partial least squares regression, PLSR) 으로 얻은 데이터와의 상관관계를 이용하여 검정선을 얻었다. 이 방법의 직선성, 정량범위, 재현성 등을 검토하여 본 분석법의 정확성을 검증하였다. 검정선의 상관계수는 0.9983이고 검정의 표준오차(SEC)는 0.14% 이였으며, 이 분석법의 시간별 및 날자별 정밀도인 CV 값은 각 각 0.45%과 0.56% 이다.

고분자 전해질 연료전지용 막가습기의 상대습도 추정을 위한 소프트센서 개발 (Soft Sensor Development for Predicting the Relative Humidity of a Membrane Humidifier for PEM Fuel Cells)

  • 한인수;신현길
    • 한국수소및신에너지학회논문집
    • /
    • 제25권5호
    • /
    • pp.491-499
    • /
    • 2014
  • It is important to accurately measure and control the relative humidity of humidified gas entering a PEM (polymer electrolyte membrane) fuel cell stack because the level of humidification strongly affects the performance and durability of the stack. Humidity measurement devices can be used to directly measure the relative humidity, but they cost much to be equipped and occupy spaces in a fuel cell system. We present soft sensors for predicting the relative humidity without actual humidity measuring devices. By combining FIR (finite impulse response) model with PLS (partial least square) and SVM (support vector machine) regression models, DPLS (dynamic PLS) and DSVM (dynamic SVM) soft sensors were developed to correctly estimate the relative humidity of humidified gases exiting a planar-type membrane humidifier. The DSVM soft sensor showed a better prediction performance than the DPLS one because it is able to capture nonlinear correlations between the relative humidity and the input data of the soft sensors. Without actual humidity sensors, the soft sensors presented in this work can be used to monitor and control the humidity in operation of PEM fuel cell systems.

Comparative Investigation of Flavors in Cigarettes by Electronic Nose and GC/MS

  • Lee, Yelin;Park, Jin-Won;Lee, Hwan-Woo;Lee, Seung-Yong;Lee, Hyung-Suk
    • 한국연초학회지
    • /
    • 제35권1호
    • /
    • pp.20-27
    • /
    • 2013
  • An Electronic Nose(E-Nose) and Gas Chromatography/Mass Spectroscopy (GC/MS) are meanwhile conventional technique to analyze volatile materials in many industries (e.g., food, medicine, environment) and have broad acceptance in the analysis of tobacco products. In this study, an experiment where tin oxide gas sensor array responses and GC/MS profiles are used to characterize the volatile compounds of different cigarettes at the same time is performed and the measurements of two instruments are compared for cigarette samples with a known chemical information. E-Nose and GC/MS were employed to differentiate and match flavored cigarettes with commercial tobacco flavoring agents (lavender, vanilla, peppermint, orange, star anise). For verifying reliability of two systems, the analyses were conducted in terms of amount of flavors in each cigarettes using partial least squares (PLS) and with the principal components analysis (PCA). Various chemical sensors and GC/MS data was reduced into two principal factors (PC1, PC2) for being distinguished with visualized regions. Both systems provided adequate results for odor characteristics of cigarettes in this study with each instrument having its own advantages and disadvantages.

Discrimination between Artemisia princeps and Artemisia capillaris Based on Near Infrared Spectroscopy Combined Multivariate Analysis

  • Lee, Dong-Young;Jeon, Min-Ji;Suh, Young-Bae;Kim, Seung-Hyun;Kim, Young-Choong;Sung, Sang-Hyun
    • Journal of Pharmaceutical Investigation
    • /
    • 제41권6호
    • /
    • pp.377-380
    • /
    • 2011
  • The Artemisia princeps (Compositae) has been used in traditional Korean medicine for the treatment of microbial infections and inflammatory diseases. Since A. princeps is generally difficult to be discriminated from A. capillaris, A. caplillaris has been misused in place of A. princeps. To solve this problem, a rapid and nondestructive method for discrimination of A. princeps and A. capillaris samples was developed using near infrared spectroscopy (NIRS) in the present study. A principal component analysis (PCA) and a partial least squares discrimination analysis (PLS-DA) were performed to discriminate two species. As a result, with the use of PLS-DA, A. princeps and A. capillaris were clustered according to their genus. These outcomes indicated that the NIRS could be useful for the discrimination between Artemisia princeps and Artemisia capillaris.

적외선 분광분석과 다변량 통계에 기반한 바이오디젤 품질분석 (Analysis of biodiesel quality based on infrared spectroscopy and multivariate statistics)

  • 김혜실;조현우;유준
    • 분석과학
    • /
    • 제25권4호
    • /
    • pp.214-222
    • /
    • 2012
  • ASTM (American Society for Testing and Materials) D6751-10은 바이오디젤의 품질 규격 뿐 아니라 분석방법 또한 제시하고 있다. 하지만 ASTM 표준에 따른 바이오디젤 및 포함된 여러 불순물의 품질 분석은 경제적, 시간적으로 부담이 크다. 본 연구는 적외선 분광분석법(infrared spectroscopy)과 다변량 통계분석법 중 하나인 PLS (partial least square method)를 이용하여 1회 측정만으로 바이오 디젤 및 불순물들의 농도를 분석하는 시스템을 개발하고자 하였다. 특히, 적외선을 이용한 분석에서 생기는 각 물질의 스펙트럼에 대한 산란 보정, 노이즈 감소 등을 위해 SNV, MSC, OSC, Savitzky-Golay 등의 4가지 전처리 방법의 성능을 비교하였다. 품질 분석에 필요한 바이오 디젤 검량 모델을 PLS로 모델링 결과, Savitzky-Golay 전처리를 하였을 때 정확도가 가장 우수함을 알았다.

Whistleblowing Intention: Theory of Planned Behavior Perspectives

  • WAHYUNI, Lili;CHARIRI, Anis;YUYETTA, Etna Afri
    • The Journal of Asian Finance, Economics and Business
    • /
    • 제8권1호
    • /
    • pp.335-341
    • /
    • 2021
  • This study aims to document empirically the individual factors that influence the intention to do whistleblowing. This study uses several variables, including internal locus of control, external locus of control, and whistleblowing intention. The use of the theory of Planned Behavior in this study is to explain and analyze the perception of behavior control as a determinant of whistleblowing intention. A quantitative research approach is used. The type of data in this study is primary data in the form of a questionnaire. The data collection method in this research is using the survey method. The sampling technique used a nonprobability sampling method, namely, the census method. The census method is the entire population sampled. The population in this study was all employees of the Pratama tax office in West Semarang. The research was conducted by distributing 111 questionnaires. Ninety-one valid questionnaires were returned appropriate for analysis. The data were processed using Partial Least Square-Structural Equation Modeling ((PLS-SEM) using the Warp PLS 7.0 program. WarpPLS 7.0 was used to test hypotheses and the relationship between variables. The study results showed that both internal locus of control and external locus of control affect whistleblowing intention.

확장된 깊이-우선 탐색 알고리듬을 적용한 다중표적 위치 좌표 추정 기법 (Location Estimation for Multiple Targets Using Expanded DFS Algorithm)

  • 박소령;노상욱
    • 한국통신학회논문지
    • /
    • 제38C권12호
    • /
    • pp.1207-1215
    • /
    • 2013
  • 이 논문에서는 장애물이 존재하는 환경에서 적외선 센서를 가진 다수의 감시 로봇이 획득한 정보를 융합하여 분산되어있는 표적의 위치 좌표를 추정하는 기법을 제안한다. 방위각(azimuth)과 표적을 대응시키는 방법으로는 장애물이 존재하지 않는 경우에서 제안되었던 깊이-우선(depth-first) 트리 탐색(tree search) 기법을 바탕으로, 우회경로 탐색, 중간 단계 탐색 종료, 하위 단계 부분 탐색, 결정기준 보완 등을 추가함으로써 트리 탐색을 확장한 새로운 기법을 제시하였다. 방위각과 표적이 대응된 후에는 하나의 표적을 가리키는 방위각들에 최소 제곱 오차(least square error) 알고리듬을 적용하여 최적 교점을 구함으로써 표적의 위치 좌표를 추정한다. 제안한 위치 추정 기법의 좌표 추정 성능과 복잡도를 모의실험으로 제시하고 분석한다.