Multivariate Calibration Transfer Employing Variable Selection and Subagging
نویسندگان
چکیده
Neste artigo, é proposta uma nova técnica para transferência de calibração, que combina o Algoritmo das Projeções Sucessivas (APS) para seleção de variáveis robustas com a técnica de sub-amostragem e agregação de modelos conhecida como subagging. A técnica proposta tem por objetivo construir modelos de Regressão Linear Múltipla (RLM) que sejam robustos com respeito a diferenças na resposta instrumental de dois espectrômetros (primário e secundário). Para isso, um pequeno conjunto de amostras de transferência, com espectros adquiridos no instrumento secundário, é empregado para guiar o procedimento de seleção de variáveis. A eficiência da técnica proposta é demonstrada em um estudo de caso envolvendo a determinação por FT-IR da massa específica e duas temperaturas de destilação (T10%, T90%) em amostras de gasolina e a determinação por NIR de umidade em amostras de milho. Em termos do erro quadrático médio de predição no espectrômetro secundário, os modelos RLM gerados pela abordagem APS-subagging forneceram resultados melhores que os obtidos por Mínimos Quadrados Parciais empregando Padronização Direta por Partes. Em particular, o uso de subagging resultou em uma redução mais sistemática do erro de predição com a inclusão progressiva de amostras de transferência.
منابع مشابه
Bagging for robust non-linear multivariate calibration of spectroscopy
This paper presents the application of the bagging technique for non-linear regression models to obtain more accurate and robust calibration of spectroscopy. Bagging refers to the combination of multiple models obtained by bootstrap re-sampling with replacement into an ensemble model to reduce prediction errors. It is well suited to “non-robust” models, such as the non-linear calibration method...
متن کاملApplication of Genetic Algorithms for Pixel Selection in MIA-QSAR Studies on Anti-HIV HEPT Analogues for New Design Derivatives
Quantitative structure-activity relationship (QSAR) analysis has been carried out with a series of 107 anti-HIV HEPT compounds with antiviral activity, which was performed by chemometrics methods. Bi-dimensional images were used to calculate some pixels and multivariate image analysis was applied to QSAR modelling of the anti-HIV potential of HEPT analogues by means of multivariate calibration,...
متن کاملApplication of Genetic Algorithms for Pixel Selection in MIA-QSAR Studies on Anti-HIV HEPT Analogues for New Design Derivatives
Quantitative structure-activity relationship (QSAR) analysis has been carried out with a series of 107 anti-HIV HEPT compounds with antiviral activity, which was performed by chemometrics methods. Bi-dimensional images were used to calculate some pixels and multivariate image analysis was applied to QSAR modelling of the anti-HIV potential of HEPT analogues by means of multivariate calibration,...
متن کاملA GPU-Based Implementation of the Firefly Algorithm for Variable Selection in Multivariate Calibration Problems
Several variable selection algorithms in multivariate calibration can be accelerated using Graphics Processing Units (GPU). Among these algorithms, the Firefly Algorithm (FA) is a recent proposed metaheuristic that may be used for variable selection. This paper presents a GPU-based FA (FA-MLR) with multiobjective formulation for variable selection in multivariate calibration problems and compar...
متن کاملMulti-Objective Evolutionary Algorithm NSGA-II for Variables Selection in Multivariate Calibration Problems
This paper proposes a multiobjective formulation for variable selection in multivariate calibration problems in order to improve the generalization ability of the calibration model. The authors applied this proposed formulation in the multiobjective genetic algorithm NSGA-II. The formulation consists in two conflicting objectives: minimize the prediction error and minimize the number of selecte...
متن کامل