Kernel partial diagnostic robust potential to handle high-dimensional and irregular data space on near infrared spectral data

Abstract

In practice, the collected spectra are very often composes of complex overtone and many overlapping peaks which may lead to misinterpretation because of its significant nonlinear characteristics. Using linear solution might not be appropriate. In addition, with a high-dimension of dataset due to large number of observations and data points the classical multiple regressions will neglect to fit. These complexities commonly will impact to multicollinearity problem, furthermore the risk of contamination of multiple outliers and high leverage points also increases. To address these problems, a new method called Kernel Partial Diagnostic Robust Potential (KPDRGP) is introduced. The method allows the nonlinear solution which maps nonlinearly the original input X matrix into higher dimensional feature mapping with corresponds to the Reproducing Kernel Hilbert Spaces (RKHS). In dimensional reduction, the method replaces the dot products calculation of elements in the mapped data to a nonlinear function in the original input space. To prevent the contamination of the multiple outlier and high leverage points the robust procedure using Diagnostic Robust Generalized Potentials (DRGP) algorithm was used. The results verified that using the simulation and real data, the proposed KPDRGP method was superior to the methods in the class of non-kernel and some other robust methods with kernel solution.

Teaching Quality Evaluation Model of a Flipped Classroom in Colleges and Universities Based on Support Vector Machine

“…In the case of nonlinearity, x is transformed into high-dimensional feature space through nonlinearity, so as to find the optimal classification surface. In the highdimensional space, the inner product operation is carried out by using the function of the original space, and the nonlinear problem in the linear space [32] is mapped to the high-dimensional feature space, so as to fundamentally solve the nonlinear problem. Therefore, assuming that the kernel function satisfying the Mercer condition is Kðx i , x j Þ, the linear classification can be obtained through nonlinear transformation without increasing the computational complexity.…”

Multivariate analysis of NMR‐based metabolomic data

“…PLS discriminant analysis (PLS-DA) is widely used for analysis of metabolomic data, and is a variant of PLS used for classification problems, ie when y is categorical. 49 The importance of variables may be assessed through the loadings and variable importance scores such as variable importance in projection (VIP) scores and the selectivity ratio (SR), 48,50 which summarize the contributions of the original variables to the model. However, PLS models are prone to overfitting and the number of latent variables should be carefully selected by validation, which is described in Section 7.…”

Automated Fitting Process Using Robust Reliable Weighted Average on Near Infrared Spectral Data Analysis

“…The 100(IV)% of the predictor variables were randomly selected as relevant variables, and the remaining were considered as less relevant. The formulation of this simulation can be defined as follows: 5,20) ( j e = 1, 2, . .…”

“…However, despite having these benefits, several studies have reported its weakness due to its robustness. The fitted model performs poorly when outliers and leverage points are present in a dataset [17,18], as it fails to fit the nonlinear behavior in the input space [19,20]. In addition, the contamination of irrelevant variables involves during the fitting process [21][22][23] is a popular topic in most discussions.…”