Search

The exploration and analyses of chemical components in (extra‐)terrestrial geological materials (such as asteroids and meteorites) are insightful in modern research. Laser‐induced breakdown spectroscopy (LiBS) is a popular method for analyzing the chemical attributes of geologic samples—which scientists use to study and understand planetary bodies and their complex histories. In the literature, several machine learning models that produce high‐accuracy predictions have been proposed. In our work, we compared the performances of such models in predicting elemental abundances on a certain spectroscopic dataset. Models included partial least squares (PLS), extreme gradient boost machines (XGB), neural networks, and linear models. In our results, we showed how PLS and XGB are superior in terms of high predictive power, their ability to generalize, and their reasonably efficient runtimes. In addition, we proposed Ensemble models that aggregate predictions of top‐tier models and observed that they can be desirable. We intend to gain better understanding of how these models perform in predicting elemental compositions on specific spectrum (LIBS) datasets. In this paper, we compare several machine learning models to identify those who performed best in a specific spectroscopic datasets. The models investigated include partial least squares, Lasso regression, support vector regression, artificial and convolutional neural networks, and extreme gradient boosting machines. We then propose an Ensemble model aggregating the predictions of the top‐tier, models and we have found that it performs significantly well. [ABSTRACT FROM AUTHOR]

Beta regression has become a popular tool for performing regression analysis on chemical, environmental, or biological data in which the dependent variable is restricted to the interval [0, 1]. For the first time, in this paper, we propose a Liu estimator for the beta regression model with fixed dispersion parameter that may be used in several realistic situations when the degree of correlation among the regressors differs. First, we show analytically that the new estimator outperforms the maximum likelihood estimator (MLE) using the mean square error (MSE) criteria. Second, using a 'simulation study, we investigate the properties in finite samples of six different suggested estimators of the shrinkage parameter and compare it with the MLE. The simulation results indicate that in the presence of multicollinearity, the Liu estimator outperforms the MLE uniformly. Finally, using an empirical application on chemical data, we show the benefit of the new approach to applied researchers. Beta regression is popular when the dependent variable is restricted to the interval [0, 1], such as rates and proportions. This paper suggests Liu estimators that may be used in the presence of multicollinearity. We show that the Liu estimator outperforms the MLE in a simulation study. Finally, using an empirical application on chemical data, we show the benefit of the new approach. [ABSTRACT FROM AUTHOR]

Transfer learning (TL), the sub‐discipline of machine learning devoted to learning from different domains, has gained increasing attention over the past decade. With the current contribution, we aim at giving a concise overview on theory, concepts, and applications of TL from a chemometrician's perspective and draw some connections to previous work on calibration model updating/adaptation and calibration transfer. Furthermore, we provide a demonstration of the application of TL in analytical chemistry and discuss the benefits and challenges associate with its use for real‐world problems. We conclude the paper by discussing some open problems and by contemplating on future research directions. [ABSTRACT FROM AUTHOR]

Orthogonal projections to latent structures (OPLS) and target projection (TP) are two alternative methods for separation of predicting and non-predicting parts of the predictor matrix in partial least squares regression (PLSR), which can also be applied on principal component regression (PCR). An additional new method called PLSO is developed in the paper. In all three methods, the predicting score vector is a scaled version of the fitted response vector. Otherwise, the resulting score and loading vectors are different, although OPLS and TP are identical within similarity transformations. All these relations are here found by simple projections of the fitted predictor matrix from PLSR or PCR, and the similarities and differences are illustrated in discriminant analysis examples. Copyright © 2014 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Laser‐induced breakdown spectroscopy (LIBS) is a new type of elemental analytical technology with the advantages of real‐time, online, and noncontact as well as enabling the simultaneous analysis of multiple elements. It has become a frontier analytical technique in spectral analysis. However, the issue of how to improve the accuracy of qualitative and quantitative analyses by extracting useful information from a large amount of complex LIBS data remains the main problem for the LIBS technique. Chemometrics is a chemical subdiscipline of multi‐interdisciplinary methods; it offers advantages in data processing, signal analysis, and pattern recognition. It can solve some complicated problems that are difficult for traditional chemical methods. In this paper, we reviewed the research progress of chemometrics methods in LIBS for spectral data preprocessing as well as for qualitative and quantitative analyses in the most recent 5 years (2012‐2016). Laser‐induced breakdown spectroscopy (LIBS) is a new type of elemental analytical technology with the advantages of without complex sample preparation, rapid and remote detection. However, the issue of how to improve the accuracy of qualitative and quantitative analyses by chemometrics methods remains the main problem for LIBS technique. The research progress of chemometrics methods in LIBS was reviewed in the most recent 5 years (2012‐2016). [ABSTRACT FROM AUTHOR]