Support vector machine classification for determination of geographical origin of Chinese ginseng using microwave plasma torch-atomic emission spectrometry

The geographical origin of Chinese ginseng is of great concern to customers, since quality varies tremendously with geographical origin. Therefore, accurately distinguishing the region of origin of specific types of ginseng, in order to differentiate the quality, is of great significance. In this paper, MPT-AES integrated with support vector machine (SVM) was proposed and applied to determine and classify the geographical origin of ginseng samples by using the chemical elemental compositions obtained. Specific data sets were extracted and dimensions were reduced through wavelet transformation. A classification model was built, relying on training sets, and then two parameters (c and g) were optimized in the SVM approach. SVM and Gaussian process classification (GPC) models were evaluated entirely on their prediction accuracy for unknown ginseng samples. Under optimized conditions, SVM outperformed GPC with a prediction accuracy of 100%, compared to 97.41%, in distinguishing the geographical origins. SVM also proved valid in the classification of individual types of ginseng with 99.81% accuracy, compared to GPC with 71.67%. These advanced chemometrics worked well for American ginseng identification. This study illustrates that chemometrics, together with the MPT-AES spectrochemical method, is a helpful and innovative technique for identifying and classifying ginseng samples, and is promising for accurate, convenient, automatic and reliable analysis.

[1]  Jan Hannig,et al.  Support vector machine classification of suspect powders using laser‐induced breakdown spectroscopy (LIBS) spectral data , 2012 .

[2]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[3]  Shouqin Zhang,et al.  Effects of ultrahigh pressure extraction conditions on yields and antioxidant activity of ginsenoside from ginseng , 2009 .

[4]  B. Villegas,et al.  Prediction of the identity of fats and oils by their fatty acid, triacylglycerol and volatile compositions using PLS-DA , 2010 .

[5]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[6]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[7]  W. Jin,et al.  High-speed photographic analysis of microwave plasma torch source behaviour , 2016 .

[8]  R. V. van Breemen,et al.  Use of high-performance liquid chromatography-tandem mass spectrometry to distinguish Panax ginseng C. A. Meyer (Asian ginseng) and Panax quinquefolius L. (North American ginseng). , 2000, Analytical chemistry.

[9]  Guanghui Niu,et al.  Classification of iron ores by laser-induced breakdown spectroscopy (LIBS) combined with random forest (RF) , 2015 .

[10]  Klaus Danzer,et al.  Determination of trace elements in wines and classification according to their provenance , 2004, Analytical and bioanalytical chemistry.

[11]  D. Sihachakr,et al.  Development of core subset for the collection of Chinese cultivated eggplants using morphological-based passport data , 2008, Plant Genetic Resources.

[12]  Karl-Heinz Ott,et al.  Metabonomics classifies pathways affected by bioactive compounds. Artificial neural network classification of NMR spectra of plant extracts. , 2003, Phytochemistry.

[13]  Xin Liu,et al.  Determining the geographical origin of Chinese green tea by linear discriminant analysis of trace metals and rare earth elements: Taking Dongting Biluochun as an example , 2016 .

[14]  Young-Ah Woo,et al.  Classification of cultivation area of ginseng by near infrared spectroscopy and ICP-AES , 2002 .

[15]  Y. Duan,et al.  Classification of steel materials by laser-induced breakdown spectroscopy coupled with support vector machines. , 2014, Applied optics.

[16]  E. Carasek,et al.  A new method of microvolume back-extraction procedure for enrichment of Pb and Cd and determination by flame atomic absorption spectrometry. , 2002, Talanta.