Liu Yisen, Zhou Songbin, Han Wei, Li Chang, Liu Weixin, Qiu Zefan, Chen Hong
Guangdong Key Laboratory of Modern Control Technology, Guangdong Academy of Sciences, Institute of Intelligent Manufacturing, Guangzhou 510070, China.
Foods. 2021 Apr 6;10(4):785. doi: 10.3390/foods10040785.
Adulteration in dairy products has received world-wide attention, and at the same time, near infrared (NIR) spectroscopy has proven to be a promising tool for adulteration detection given its advantages of real-time response and non-destructive analysis. Regardless, the accurate and robust NIR model for adulteration detection is hard to achieve in practice. Convolutional neural network (CNN), as a promising deep learning architecture, is difficult to apply to such chemometrics tasks due to the high risk of overfitting, despite the breakthroughs it has made in other fields. In this paper, the ensemble learning method based on CNN estimators was developed to address the overfitting and random initialization problems of CNN and applied to the determination of two infant formula adulterants, namely hydrolyzed leather protein (HLP) and melamine. Moreover, a probabilistic wavelength selection method based on the attention mechanism was proposed for the purpose of finding the best trade-off between the accuracy and the diversity of the sub-models in ensemble learning. The overall results demonstrate that the proposed method yielded superiority regression performance over the comparison methods for both studied data sets, and determination coefficients (R) of 0.961 and 0.995 were obtained for the HLP and the melamine data sets, respectively.
乳制品掺假问题已受到全球关注，与此同时，近红外（NIR）光谱因其具有实时响应和无损分析的优点，已被证明是一种用于掺假检测的有前途的工具。尽管如此，用于掺假检测的准确且稳健的近红外模型在实践中却难以实现。卷积神经网络（CNN）作为一种有前途的深度学习架构，尽管在其他领域取得了突破，但由于过拟合风险高，很难应用于此类化学计量学任务。本文开发了基于CNN估计器的集成学习方法，以解决CNN的过拟合和随机初始化问题，并将其应用于两种婴儿配方奶粉掺假物即水解皮革蛋白（HLP）和三聚氰胺的测定。此外，为了在集成学习中找到子模型的准确性和多样性之间的最佳平衡，提出了一种基于注意力机制的概率波长选择方法。总体结果表明，对于两个研究数据集，所提出的方法在回归性能上优于比较方法，并且HLP和三聚氰胺数据集的决定系数（R）分别为0.961和0.995。
