BACKGROUND The acid value is one of the significant indicators for evaluating the quality of edible oil during storage. Herein, this study employs a portable near-infrared spectroscopy (NIRS) system to determine the acid value during edible oil storage. Four MPA-based variable selection methods, namely, competitive adaptive reweighted sampling (CARS), variable iterative space shrinkage approach (VISSA), iteratively variable subset optimization (IVSO) and bootstrapping soft shrinkage (BOSS), were introduced to optimize the preprocessed NIR spectra. Then, support vector machine (SVM) models based on characteristic spectra obtained by different selection methods were established to achieve quantitative detection of the acid value during edible oil storage. RESULTS The results obtained revealed that compared to the full-spectrum SVM model, the SVM models established by the characteristic wavelengths optimized by the variable selection methods based on the MPA strategy exhibit a significant improvement in complexity and generalization performance. Furthermore, compared with the CARS, VISSA and IVSO methods, the BOSS method obtained the least number of characteristic wavelength variables, and the SVM model established based on the optimized features of this method exhibited the best prediction performance, with RMSEP = 0.11 mg g-1 , RP 2 =0.92 and RPD=2.82, respectively. CONCLUSION The overall results indicate that the variable selection methods based on the MPA strategy can select more targeted characteristic variables, which has good application prospects in NIR spectra feature optimization. This article is protected by copyright. All rights reserved.

中文翻译：

---

使用便携式 NIR 光谱系统结合基于 MPA 策略的变量选择算法测定食用油储存期间的酸值

背景技术酸值是评价食用油贮藏质量的重要指标之一。在此，本研究采用便携式近红外光谱 (NIRS) 系统来确定食用油储存过程中的酸值。引入了四种基于 MPA 的变量选择方法，即竞争性自适应重加权采样 (CARS)、可变迭代空间收缩方法 (VISSA)、迭代可变子集优化 (IVSO) 和引导软收缩 (BOSS) 来优化预处理的 NIR 光谱. 然后，基于不同选择方法获得的特征谱建立支持向量机（SVM）模型，实现食用油贮藏过程中酸值的定量检测。结果 获得的结果表明，与全谱 SVM 模型相比，通过基于MPA策略的变量选择方法优化的特征波长建立的SVM模型在复杂度和泛化性能上都有显着的提高。此外，与CARS、VISSA和IVSO方法相比，BOSS方法获得的特征波长变量数最少，基于该方法优化特征建立的SVM模型预测性能最好，RMSEP = 0.11 mg g- 1，RP 2 分别=0.92和RPD=2.82。结论总体结果表明，基于MPA策略的变量选择方法可以选择更有针对性的特征变量，在近红外光谱特征优化中具有良好的应用前景。本文受版权保护。版权所有。