Near-infrared spectroscopy (NIRS) combined with two discriminative analysis methods including self-organization mapping (SOM) and support vector machine (SVM) were used for discriminant analysis of vinegar with different production year and brand. Continuous wavelet transform (CWT) was adopted for spectra preprocessing. Principal component analysis (PCA) was used for spectra dimension reduction and space distribution analysis. The results shown that CWT can effective eliminate spectra translation error. PCA can greatly reduce characteristic spectrum variables and improve modeling efficiency. For identification of vinegar with different production year, the CWT-PCA-SOM method can get 97.37% correct recognition rate (CRR), and the CWT-PCA-SVM method can get 100% CRR. For identification of vinegar brand, the CWT-PCA-SOM and CWT-PCA-SVM methods can obtain 100% CRR. Near-infrared spectroscopy combined with CWT-PCA-SOM and CWT-PCA-SVM methods can both obtain better analysis results for identification of vinegar with different production year and brand, and this method has good application prospect.
Yong, HAO; Xiang, ZHAO; Qinhua, WEN; and Bin, CHEN
"Research on qualitative analysis of vinegar by using near-infrared spectroscopy combined with SOM and SVM,"
Food and Machinery: Vol. 32:
5, Article 11.
Available at: https://www.ifoodmm.cn/journal/vol32/iss5/11
 Pizarro C, Esteban-Diez I, Saenz-Gonzalez C, et al. Vinegar classfication based on feature extraction and selection from headspace solid-phase microextraction/gas chromatography volatile analyses: A feasibility study[J]. Analytica Chimica Acta, 2008, 608(1): 38-47.
 刘杨岷, 张家骊, 王利平, 等. 食醋风味成分比较研究[J]. 食品与机械, 2005, 21(5): 40-43.
 Ubeda C, Callejon R M, Hidalgo C, et al. Determination of major volatile compounds during the production of fruit vinegars by static headspace gas chromatography-mass spectrometry method[J]. Food Research International, 2011, 44(1): 259-268.
 Callejon R M, Amigo J M, Pairo E, et al. Classification of Sherry vinegars by combining multidimensional fluorescence, parafac and different classification approaches[J]. Talanta, 2012, 88(15): 456-462.
 Boffoa E F, Tavaresa L A, Ferreira M M C, et al. Classification of Brazilian vinegars according to their 1H NMR spectra by pattern recognition analysis[J]. LWT - Food Science and Technology, 2009, 42(9): 1 455-1 460.
 Zhang Qin-yi, Zhang Shun-ping, Xie Chang-sheng, et al. Characterization of Chinese vinegars by electronic nose[J]. Sensors and Actuators B: Chemical, 2006, 119(2): 538-546.
 Liu Fei, He Yong, Wang Li. Determination of effective wavelengths for discrimination of fruit vinegars using near infrared spectroscopy and multivariate analysis[J]. Analytica Chimica Acta, 2008, 615(1): 10-17.
 Cocchi M, Durante C, Foca G, et al. Application of a wavelet-based algorithm on HS—SPME/GC signals for the classification of balsamic vinegars[J]. Chemometrics and Intelligent Laboratory Systems, 2004, 71(2): 129-140.
 Hsieh Chang-wei, Li Po-hsien, Cheng Ju-yun, et al. Using SNIF—NMR method to identify the adulteration of molasses spirit vinegar by synthetic acetic acid in rice vinegar[J]. Industrial Crops and Products, 2013, 50: 904-908.
 Yip W L, Soosainather T C, Dyrstad K, et al. Classification of structurally related commercial contrast media by near infrared spectroscopy[J]. Journal of Pharmaceutical and Biomedical Analysis, 2014, 90(5):148-160.
 Coppa M, Revello-Chion A, Giaccone D, et al. Comparison of near and medium infrared spectroscopy to predict fatty acid composition on fresh and thawed milk[J]. Food Chemistry, 2014, 150(1): 49-57.
 杨代明, 方宣启. 非线性化学指纹图谱技术在食醋鉴别中的应用研究[J]. 食品与机械, 2014, 30(5): 68-71.
 Casale M, Abajo M J S, Saiz J M G, et al. Study of the aging and oxidation processes of vinegar samples from different origins during storage by near-infrared spectroscopy[J]. Analytica Chimica Acta, 2006, 557(2): 360-366.
 Liu Fei, He Yong, Wang Li. Comparison of calibrations for the determination of soluble solids content and pH of rice vinegars using visible and short-wave near infrared spectroscopy[J]. Analytica Chimica Acta, 2008, 610(2): 196-204.
 Saiz-Abajo M J, Gonzalez-Saiz J M, Pizarro C. Prediction of organic acids and other quality parameters of wine vinegar by near-infrared spectroscopy: A feasibility study[J]. Food Chemistry, 2006, 99(3): 615-621.
 Chen Quan-sheng, Ding Jiao, Cai Jian-rong, et al. Rapid measurement of total acid content (TAC) in vinegar using near infrared spectroscopy based on efficient variables selection algorithm and nonlinear regression tools[J]. Food Chemistry, 2012, 135(2): 590-595.
 夏蓉, 郝勇. 近红外光谱在食醋品牌和贮藏年份鉴别中的应用研究[J]. 中国酿造, 2012, 31(11): 27-29.
 Shao Xue-guang, Ma Chao-xiong. A general approach to derivative calculation using wavelet transform[J]. Chemometrics and Intelligent Laboratory Systems, 2003, 69(2): 157-165.
 Leung A K M, Chau F T, Gao J B. Wavelet transform: A novel method for derviative calculation in analytical chemistry[J]. Analytical Chemistry, 1998, 70(2): 5 222-5 229.
 Yan Ai-xia, Nie Xiang-lei, Wang Kai, et al. Classification of Aurora kinase inhibitors by self-organizing map (SOM) and support vector machine (SVM) [J]. European Journal of Medicinal Chemistry, 2013, 61: 73-83.
 Merlin P, Sorjamaa A, Maillet B, et al. X-SOM and L-SOM: A double classification approach for missing value imputation[J]. Neurocomputing, 2010, 73(9): 1 103-1 108.
 Nakayama N, Oketani M, Kawamura Y, et al. Classification of acute liver failure of indeterminate etiology: usefulness of clustering analysis using a self-organizing map (SOM)[J]. Journal of Hepatology, 2011, 54(1): S369-S370.
 Devos O, Downey G, Duponchel L. Simultaneous data pre-processing and SVM classification model selection based on a parallel genetic algorithm applied to spectroscopic data of olive oils[J]. Food Chemistry, 2014, 148(1): 124-130.
 Wang Ya-sheng, Yang Meng, Wei Gao, et al. Improved PLS regression based on SVM classification for rapid analysis of coal properties by near-infrared reflectance spectroscopy[J]. Sensors and Actuators B: Chemical, 2014, 193(2): 723-729.
 Sugumaran V, Ramachandran K. Effect of number of features on classification of roller bearing faults using SVM and PSVM[J]. Expert Systems with Applications, 2011, 38(4): 4 088-4 096.
 Mu T T, Nandi A K. Breast cancer detection from FNA using SVM with different parameter tuning systems and SOM-RBF classifier[J]. Journal of the Franklin Institute, 2007, 344(3/4): 285-311.
 Norinder U, Ek M E. QSAR investigation of NaV1.7 active compounds using the SVM/Signature approach and the Bioclipse Modeling platform[J]. Bioorganic & Medicinal Chemistry Letters, 2013, 23(1): 261-263.