Study on the classification and identification of various carbonate and sulfate mineral medicines based on Raman spectroscopy combined with PCA-SVM algorithm

• Published in: Analytical sciences Vol. 39; no. 2; pp. 241 - 248
• Main Authors: Han, Siqingaowa, Jin, Zhu, Deji, Dema, Han, Tana, Zhang, Yulan, Feng, Meiling, Hasi, Wuliji
• Format: Journal Article
• Language: English
• Published: Singapore Springer Nature Singapore 01.02.2023
• Subjects: Algorithms; Analytical Chemistry; Chemistry; Chemistry and Materials Science; Original Paper; Principal Component Analysis; Spectrum Analysis, Raman - methods; Sulfates; Support Vector Machine; Support vector machine (SVM); Raman spectroscopy; Carbonate mineral medicines; Sulfate mineral medicines; Principal component analysis (PCA); Classification and identification
• ISSN: 0910-6340; 1348-2246; 1348-2246
• DOI: 10.1007/s44211-022-00224-1

The efficacy of mineral medicines varies greatly between different origins. Therefore, investigating a method to quickly identify similar mineral medicines is meaningful. In this paper, a visual classification and identification model of Raman spectroscopy combined with principal component analysis (PCA) and support vector machine (SVM) algorithms was developed to rapidly classify and identify carbonate and sulfate mineral medicines. The results reveal that although the Raman spectra are too similar to distinguish by naked eye, the PCA-SVM algorithm can perform accurate classification and identification, and its accuracy, precision, recall and F1-score parameters all reach 100%. The proposed method is rapid, accurate, nondestructive, convenient, portable, and low cost, and has important application value for the classification, identification and quality supervision of various carbonate and sulfate mineral medicines. Graphical Abstract