Related Experiment Video
Updated: Aug 5, 2026
07:15
Measuring Interactions between Fluorescent Probes and Lignin in Plant Sections by sFLIM Based on Native Autofluorescence
Published on: January 2, 2020
Classification of fermentation methods for white mulberry products using fluorescence spectroscopy combined with deep
Tianrui Zeng1,2, Hao Yang3,2, Shimu Wang3,2
1College of Light Industry Science and Engineering, Beijing Technology and Business University, Beijing, 100048, China.
Analytical Methods : Advancing Methods and Applications
|July 27, 2026
Summary
Identifying cosmetic raw materials like plant fermentation extracts is key for quality. Deep learning, particularly the Informer model, excels at distinguishing these extracts using fluorescence spectroscopy, offering superior accuracy over traditional methods.
Area of Science:
- Cosmetic Science
- Analytical Chemistry
- Biotechnology
Background:
- Accurate identification of cosmetic raw materials, such as plant fermentation extracts versus water extracts, is vital for product quality, safety, and efficacy.
- Conventional chemometrics and existing machine learning methods struggle with the nonlinear characteristics of fluorescence spectra and exhibit limitations in feature representation and generalization.
Purpose of the Study:
- To develop and evaluate advanced methods for distinguishing between plant fermentation extracts and water extracts in cosmetic raw materials.
- To compare the performance of deep learning models against traditional chemometric and machine learning techniques for classifying these extracts.
Main Methods:
- Preparation of 966 white mulberry (Morus alba L.) samples, including yeast fermentation extracts, lactic acid bacteria fermentation extracts, and water extracts.
- Acquisition of fluorescence spectra for all prepared samples.
- Construction and comparative analysis of six deep learning models (Informer, PatchTST, Transformer, TCN, LSTM, CNN) and traditional models (SVM, PCA-LDA, PLS-DA).
Main Results:
- The Informer deep learning model achieved the highest classification performance, with accuracy, precision, recall, and F1 score of 0.981, 0.982, 0.981, and 0.981, respectively.
- The Informer model significantly outperformed other deep learning models and traditional methods (SVM, PCA-LDA, PLS-DA) in feature extraction and generalization.
- The study demonstrates the effectiveness of integrating fluorescence spectroscopy with deep learning for raw material identification.
Conclusions:
- Deep learning models, especially the Informer architecture, offer a powerful and accurate solution for identifying liquid cosmetic raw materials based on fluorescence spectra.
- This approach overcomes the limitations of conventional methods in analyzing complex spectral data.
- The findings provide a novel and effective strategy for quality control and authenticity verification in the cosmetic industry.
Related Concept Videos
01:30Methods to Assess Microbial Populations
Assessing microbial populations is crucial for understanding microbial roles in health, ecology, and industry. Various complementary techniques—both culture-based and molecular—enable detailed analysis of microbial abundance, diversity, and function.Viable Plate CountThe viable plate count is a traditional culture-based method used to estimate the number of living microbes in a sample. After serial dilution, the sample is spread onto nutrient agar plates. Each viable cell forms a visible...
01:28Methods of Classification and Identification
Bacterial identification relies on a diverse array of techniques to classify and understand microorganisms, each tailored to uncover specific characteristics. Traditional morphological approaches, while still valuable, are limited for closely related or structurally simple organisms. Modern methods integrate biochemical, serological, genetic, and advanced molecular tools to achieve greater accuracy.Morphological and Biochemical TechniquesMorphological characteristics, such as cell shape and...