Spectral characteristic variable selection and optimization method based on Stacking integrated framework

Abstract

The invention discloses a spectral characteristic variable selection and optimization method based on a Stacking integration framework, and the method comprises the steps: constructing a sample set, and dividing the sample set into a training sample set and a testing sample set; selecting a representative characteristic variable selection method from the large classes of the variable interval selection method, the variable information selection method and the variable optimization selection method; constructing a plurality of base learners, integrating the base learners by adopting a Stackingintegration framework, constructing a meta-learner, and taking the output of the base learners as the input of the meta-learner; training and testing a base learner and a meta learner of the Stackingintegrated framework by using the sample set; and inputting the to-be-detected spectrum information into the basic learner, and obtaining a detection result of the to-be-detected spectrum according tothe output of the meta-learner. According to the spectral characteristic variable selection and optimization method based on the Stacking integrated framework, the defects of a single characteristicvariable selection method are overcome, the detection precision of a test sample is high, and the stability of a detection result is good.

Description

technical field [0001] The invention belongs to the field of spectral analysis, and in particular relates to a method for selecting and optimizing spectral characteristic variables based on a Stacking integration framework. Background technique [0002] According to the definition of the American Society for Testing and Materials Testing, the NIR (Near Infrared) region of the near-infrared spectrum refers to electromagnetic waves with a wavelength in the range of 780-2526nm. It belongs to the double frequency and main frequency absorption spectrum of the molecular vibration spectrum, and the near-infrared spectrum is mainly generated when the molecular vibration transitions from the ground state to a high energy level due to the anharmonic nature of the molecular vibration. The near-infrared spectral region is consistent with the combined frequency of vibrations of hydrogen-containing groups (O-H, N-H, C-H) in organic molecules and the absorption regions of multiple levels o...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to solve the above problems, propose a spectral feature variable selection and optimization method based on the Stacking integration framework, overcome the defect of a single feature variable selection method, and improve prediction accuracy

Images