High-level Fusion Coupled with Mahalanobis Distance Weighted (MDW) Method for Multivariate Calibration

Abstract

near infrared spectra (niR) technology is a widespread detection method with high signal to noise ratio (SnR) while has poor modeling interpretation due to the overlapped features. Alternatively, midinfrared spectra (MiR) technology demonstrates more chemical features and gives a better explanation of the model. Yet, it has the defects of low SnR. With the purpose of developing a model with plenty of characteristics as well as with higher SnR, niR and MiR technologies are combined to perform highlevel fusion strategy for quantitative analysis. A novel chemometrical method named as Mahalanobis distance weighted (MDW) is proposed to integrate niR and MiR techniques comprehensively. Mahalanobis distance (MD) based on the principle of spectral similarity is obtained to calculate the weight of each sample. Specifically, the weight is assigned to the inverse ratio of the corresponding MD. Besides, the proposed MDW method is applied to niR and MiR spectra of active ingredients in deltamethrin and emamectin benzoate formulations for quantitative analysis. As a consequence, the overall results show that the MDW method is promising with noticeable improvement of predictive performance than individual methods when executing high-level fusion for quantitative analysis. Gas chromatography 1 , high-performance liquid chromatography 2 , thin-layer chromatography 3 , gas chromatography-mass spectrometry, liquid chromatography-mass spectrometry 4,5 are frequently-used techniques to detect the concentration of active ingredients in pesticides. While they may come across some inconveniences such as time-consuming, expensive lab-cost and complex pre-treatment. Admittedly, vibrational spectroscopic techniques respond fast and require no pre-treatment which are especially suitable for real-time measurement and rapid analysis 6,7. NIR and MIR techniques have been widely used to determine the active ingredients and illegal added component in pesticide formulations 8-12. The two commonly used techniques of NIR and MIR are generated by the transition of vibrational energy levels. The transition occurs only when it absorbs a certain wavelength. To be concrete, as for MIR, it mainly absorb the fundamental vibrations of C-H, N-H and O-H, whereas, NIR contains a wealthy information of the combinations and overtones of the fundamental vibrations of C-H, N-H and O-H. Data fusion is a tool to combine the data originated from different sources comprehensively. To a certain degree, a single technique can only perceive limited partial information, therefore multiple techniques are expected to provide sufficient information from different viewpoints. Data fusion techniques 13-15 have been extensively employed with the purpose of getting a more well-rounded result. Generally, data fusion strategies can be basically classified into three levels, i.e., low-level, mid-level and high-level data fusion 16. When executing high-level fusion, individual models for each data source are developed first, and afterwards the final results are obtained by combinin...

The New Approaches to Identification of Tinctures and Medicinal Plants

“…For the motherwort herb (6-10), three components (PC1-PC3) predominantly had positive values. A common feature observed in hawthorn fruits (11)(12)(13)(14)(15) from various producers was the presence of negative values for the first three components (PC1, PC2, and PC3), with a few exceptions.…”

“…The Mahalanobis distance [15] is a measure that determines the level of difference between objects. Differences are considered significant with a 95% probability if they exceed 3 standard deviations (≥3SD) from the mean value of the spectral measurement.…”

A Mahalanobis Distance-Based Approach for Dynamic Multiobjective Optimization With Stochastic Changes

“…Therefore, the technology that includes learning from previous environments and using the learned result in a new environment is key to MDA in dealing with DMOPs with stochastic changes. Besides, the MD not only considers all variable characteristics of sample sets but can also effectively deal with nonindependent and identical distribution among different variables [32], [33]. Consequently, the MD can reasonably learn from the previous environments for handling DMOPs with stochastic changes (i.e., Section III-A1).…”

“…Consequently, how to solve this kind of stochastic DMOPs is the subject of our most significant concern in this paper. Additionally, Mahalanobis distance (MD) [30] is extended from the Euclidean distance [31], which can effectively deal with the issue of non-independent and identical distribution among different dimensions or variables and consider all variable characteristics of sample sets [32], [33]. For that, MD is frequently presented in different fields to test the similarity between two independent sample sets [34].…”

“…For that, MD is frequently presented in different fields to test the similarity between two independent sample sets [34]. Inspired by [32], [33], we hypothesize that MD can detect the similarity between the sample sets from different evolutionary environments.…”