Robust NIR quantitative model using MIC-SPA variable selection and GA-ELM

In order to increase the accuracy and robustness of the calibration model, a feature selection method MIC-SPA (maximal information coefficient successive projections algorithm) combined with GA-ELM (extreme learning machine genetic algorithm) modeling method was proposed in this paper. MIC was used to extract the feature set with high correlation with the target variable, which can filter the non-informative variables and noise data in the spectrum. Then, SPA was applied to further eliminate redundant features with the maximum projection value on the orthogonal subspace of the previously selected wavelength, and the features corresponding to the minimum RMSECV were selected as the optimized feature set. While the sample set partitioning based on joint X-Y distance (SPXY) method was also introduced for increasing the diversity of the training data set, Furthermore, GA-ELM was introduced to establish the robust NIR analysis model by using the advantages of neural network for non-linear data processing. In order to investigate the effectiveness of the algorithm, the MIC-SPA was compared with the commonly used feature selection methods including least angle regression (LARS), uninformative variables elimination (UVE), competitive adaptive reweighted sampling (CARS), SPA and MIC-LARS, MIC-UVE, MIC-CARS respectively, the number of selected features, predictive ability and robustness of the model were also evaluated. The results confirmed that the accuracy and robustness of NIRS model can be obtained by combining MIC-SPA and GA-ELM methods. It will valuable for the application of quantitative analysis by NIR spectroscopy.

Automated summary

This paper proposes a feature selection method MIC-SPA and GA-ELM modeling method to improve the accuracy and robustness of the calibration model. MIC is used to extract the feature set with high correlation with the target variable, filtering non-informative variables and noise data in the spectrum. SPA is applied to eliminate redundant features, and the features with the minimum RMSECV are selected as the optimized feature set. The results show that the combination of MIC-SPA and GA-ELM methods can achieve accuracy and robustness in NIRS modeling.