# Fuzzy Evaluation Output of Taste Information for Liquor Using Electronic Tongue Based on Cloud Model

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Acquisition of Liquor Taste Information

#### 2.1. Materials

#### 2.2. Collection of Liquor Taste Information Based on Electronic Tongue System

#### 2.3. Acquisition of Liquor Taste Information Based on Artificial Sensory Evaluation

## 3. Establishment of Liquor Taste Information Cloud Model

#### 3.1. Determination of Liquor Flavor Discrimination Algorithm

**g**(kernel function) and

**c**(penalty factor) to construct and solve the optimization problem [35]:

**c**; kernel function

**g**) to obtain a relatively ideal classification accuracy. In general, the idea of cross-validation (CV) can avoid the occurrence of over-learning and under-learning states, resulting in optimal parameters. In the sense of CV, the optimization of parameters by genetic algorithm (GA) can be conducive to find parameter

**c**and

**g**faster and more stable. To improve the performance of the classifier, the SVM optimized by GA(SVM(GA)) was used to classify four different types of liquor in this study, in which the maximum number of iterations was 200, the population number was 20, the search range of c was ${2}^{-10}$ to ${2}^{10}$, and the search range of g was ${2}^{-10}$ to ${2}^{10}$. In this experiment, each group of fragrant liquor was repeated 20 times. In the case where the ratio of training set to test set is 7:3, 14 groups were randomly selected as the training set and the remaining six groups were used as the prediction set in each liquor data. The classification results of the electronic tongue liquor data are shown in Figure 2.

**c**and

**g**were output. Therefore, when the highest accuracy of five cross-validation of the training set was 100%, the optimal parameters of

**c**and

**g**were separately 24.2335 and 0.00066757. On this basis, SVM prediction model was built to train and predict the liquor data, and the classification prediction results in Figure 2b were obtained. Where the blue circles and red circles represent the classification predicted by SVM and the actual category of the data, respectively. It can be seen that the prediction classification accuracy reached 100%. The results show that the SVM(GA) has a positive impact on the prediction performance of the model. In this method, different types of liquor can be predicted well. Therefore, SVM(GA) was used to discriminate liquor flavors, and the results were substituted into the cloud model to obtain the final fuzzy evaluation.

#### 3.2. The Concept of the Cloud Model

#### 3.3. Liquor Taste Information Cloud Drop Point Acquisition

#### 3.4. Frequency of Words in Liquor Taste Information

#### 3.5. Correlation between the Range of Liquor Taste Information Cloud Droplets and Evaluation Words

#### 3.5.1. Contribution of Cloud Droplet Groups to Qualitative Concepts

#### 3.5.2. Correlation of Words in the Cloud Droplet Central Areas

#### 3.5.3. Correlation of Words in the Cloud Droplet Ring Areas

#### 3.5.4. Correlation Result of Cloud Droplet Areas and Evaluation Words

## 4. Results of Fuzzy Evaluation of Liquor Flavor

## 5. Conclusions

## Author Contributions

## Funding

## Conflicts of Interest

## Appendix A

Flavor of Liquor | Area | Words |
---|---|---|

jiang-flavor style | $\frac{{(x-41.6501)}^{2}}{{2.8881}^{2}}+\frac{{(y+23.684)}^{2}}{{2.4452}^{2}}\le 1$ | fully mellow |

$\begin{array}{l}\frac{{(x-41.6501)}^{2}}{{4.1259}^{2}}+\frac{{(y+23.684)}^{2}}{{3.4932}^{2}}\le 1\\ \&\frac{{(x-41.6501)}^{2}}{{2.8881}^{2}}+\frac{{(y+23.684)}^{2}}{{2.4452}^{2}}>1\end{array}$ | elegant and delicate | |

$\begin{array}{l}\frac{{(x-41.6501)}^{2}}{{4.84}^{2}}+\frac{{(y+23.684)}^{2}}{{4.0987}^{2}}\le 1\\ \&\frac{{(x-41.6501)}^{2}}{{4.1259}^{2}}+\frac{{(y+23.684)}^{2}}{{3.4932}^{2}}>1\end{array}$ | full bodied | |

$\begin{array}{l}\frac{{(x-41.6501)}^{2}}{{6.1613}^{2}}+\frac{{(y+23.684)}^{2}}{{5.2165}^{2}}\le 1\\ \&\frac{{(x-41.6501)}^{2}}{{4.84}^{2}}+\frac{{(y+23.684)}^{2}}{{4.0987}^{2}}>1\end{array}$ | long aftertaste | |

$\begin{array}{l}\frac{{(x-41.6501)}^{2}}{{8.2518}^{2}}+\frac{{(y+23.684)}^{2}}{{6.9864}^{2}}\le 1\\ \&\frac{{(x-41.6501)}^{2}}{{6.1613}^{2}}+\frac{{(y+23.684)}^{2}}{{5.2165}^{2}}>1\end{array}$ | coordination | |

feng-flavor style | $\frac{{(x+31.8733)}^{2}}{{2.2081}^{2}}+\frac{{(y+23.6339)}^{2}}{{2.9945}^{2}}\le 1$ | sweet and cool |

$\begin{array}{l}\frac{{(x+31.8733)}^{2}}{{3.4585}^{2}}+\frac{{(y+23.6339)}^{2}}{{4.6901}^{2}}\le 1\\ \&\frac{{(x+31.8733)}^{2}}{{2.2081}^{2}}+\frac{{(y+23.6339)}^{2}}{{2.9945}^{2}}>1\end{array}$ | long clean tail | |

$\begin{array}{l}\frac{{(x+31.8733)}^{2}}{{4.6823}^{2}}+\frac{{(y+23.6339)}^{2}}{{6.3497}^{2}}\le 1\\ \&\frac{{(x+31.8733)}^{2}}{{3.4585}^{2}}+\frac{{(y+23.6339)}^{2}}{{4.6901}^{2}}>1\end{array}$ | mellow and elegant | |

$\begin{array}{l}\frac{{(x+31.8733)}^{2}}{{5.9593}^{2}}+\frac{{(y+23.6339)}^{2}}{{8.0815}^{2}}\le 1\\ \&\frac{{(x+31.8733)}^{2}}{{4.6823}^{2}}+\frac{{(y+23.6339)}^{2}}{{6.3497}^{2}}>1\end{array}$ | all tastes harmonize | |

$\begin{array}{l}\frac{{(x+31.8733)}^{2}}{{7.9812}^{2}}+\frac{{(y+23.6339)}^{2}}{{10.8234}^{2}}\le 1\\ \&\frac{{(x+31.8733)}^{2}}{{5.9593}^{2}}+\frac{{(y+23.6339)}^{2}}{{8.0815}^{2}}>1\end{array}$ | mellow fullness | |

nong-flavor style | $\frac{{(x-34.2356)}^{2}}{{2.6959}^{2}}+\frac{{(y+27.0657)}^{2}}{{2.1275}^{2}}\le 1$ | soft and sweet |

$\begin{array}{l}\frac{{(x-34.2356)}^{2}}{{4.1636}^{2}}+\frac{{(y+27.0657)}^{2}}{{3.2858}^{2}}\le 1\\ \&\frac{{(x-34.2356)}^{2}}{{2.6959}^{2}}+\frac{{(y+27.0657)}^{2}}{{2.1275}^{2}}>1\end{array}$ | sweet and refreshing | |

$\begin{array}{l}\frac{{(x-34.2356)}^{2}}{{5.8410}^{2}}+\frac{{(y+27.0657)}^{2}}{{4.6096}^{2}}\le 1\\ \&\frac{{(x-34.2356)}^{2}}{{4.1636}^{2}}+\frac{{(y+27.0657)}^{2}}{{3.2858}^{2}}>1\end{array}$ | mellow | |

$\begin{array}{l}\frac{{(x-34.2356)}^{2}}{{6.7097}^{2}}+\frac{{(y+27.0657)}^{2}}{{5.2951}^{2}}\le 1\\ \&\frac{{(x-34.2356)}^{2}}{{5.8410}^{2}}+\frac{{(y+27.0657)}^{2}}{{4.6096}^{2}}>1\end{array}$ | alcohol harmonious | |

$\begin{array}{l}\frac{{(x-34.2356)}^{2}}{{8.9862}^{2}}+\frac{{(y+27.0657)}^{2}}{{7.0917}^{2}}\le 1\\ \&\frac{{(x-34.2356)}^{2}}{{6.7097}^{2}}+\frac{{(y+27.0657)}^{2}}{{5.2951}^{2}}>1\end{array}$ | long aftertaste | |

mild-flavor style | $\frac{{(x-4.4596)}^{2}}{{2.523}^{2}}+\frac{{(y+36.482)}^{2}}{{1.6075}^{2}}\le 1$ | pure fragrance |

$\begin{array}{l}\frac{{(x-4.4596)}^{2}}{{4.0706}^{2}}+\frac{{(y+36.482)}^{2}}{{2.5934}^{2}}\le 1\\ \&\frac{{(x-4.4596)}^{2}}{{2.523}^{2}}+\frac{{(y+36.482)}^{2}}{{1.6075}^{2}}>1\end{array}$ | long aftertaste | |

$\begin{array}{l}\frac{{(x-4.4596)}^{2}}{{5.0462}^{2}}+\frac{{(y+36.482)}^{2}}{{3.2150}^{2}}\le 1\\ \&\frac{{(x-4.4596)}^{2}}{{4.0706}^{2}}+\frac{{(y+36.482)}^{2}}{{2.5934}^{2}}>1\end{array}$ | sweet and soft | |

$\begin{array}{l}\frac{{(x-4.4596)}^{2}}{{6.56}^{2}}+\frac{{(y+36.482)}^{2}}{{4.1794}^{2}}\le 1\\ \&\frac{{(x-4.4596)}^{2}}{{5.0462}^{2}}+\frac{{(y+36.482)}^{2}}{{3.2150}^{2}}>1\end{array}$ | natural coordination | |

$\begin{array}{l}\frac{{(x-4.4596)}^{2}}{{10.0923}^{2}}+\frac{{(y+36.482)}^{2}}{{6.4299}^{2}}\le 1\\ \&\frac{{(x-4.4596)}^{2}}{{6.56}^{2}}+\frac{{(y+36.482)}^{2}}{{4.1794}^{2}}>1\end{array}$ | sweet and refreshing |

## References

- Toko, K. Taste sensor. Sens. Actuators
**2000**, 64, 334–342. [Google Scholar] [CrossRef] - Toko, K. Sensors for Measuring Taste and Smell. Brain Nerve.
**2017**, 69, 557–563. [Google Scholar] [PubMed] - Zhu, N.; Zou, Y.; Huang, M.; Dong, S.; Wu, X.; Liang, G. A sensitive, colorimetric immunosensor based on Cu-MOFs and HRP for detection of dibutyl phthalate in environmental and food samples. Talanta
**2018**, 186, 104–109. [Google Scholar] [CrossRef] [PubMed] - Li, J.W.; Hou, C.J.; Huo, D.Q.; Mei, Y.; Zhang, S.Y.; Ma, Y.; Lin, Y. A minimalist Chinese liquor identification system based on a colorimetric sensor array with multiple applications. Anal. Methods
**2017**, 1, 141–148. [Google Scholar] - Xia, T.Y.; Fu, S.X.; Wang, Q.H.; Wen, Y.; Chan, S.A.; Zhu, S.; Gao, S.H.; Tao, X.; Zhang, F.; Chen, W.S. Targeted metabolomic analysis of 33 amino acids and biogenic amines in human urine by ion-pairing HPLC-MS/MS: Biomarkers for tacrolimus nephrotoxicity after renal transplantation. Biomed. Chromatogr.
**2018**, 7, e4198. [Google Scholar] [CrossRef] - Zhang, L.Q.; Yi, L.; Ying, L.; Liang, K.H.; Fang, Z.; Tao, X.; Wang, M.M.; Wang, H.X.; Song, X.J.; Lu, B.Y. Determination of phenolic acid profiles by HPLC-MS in vegetables commonly consumed in China. Food Chem.
**2019**, 276, 538–546. [Google Scholar] [CrossRef] - Jira, W.; Münch, S. A sensitive HPLC-MS/MS screening method for the simultaneous detection of barley, maize, oats, rice, rye and wheat proteins in meat products. Food Chem.
**2019**, 275, 214–223. [Google Scholar] [CrossRef] - Majchrzak, T.; Wojnowski, W.; Dymerski, T.; Gębicki, J.; Namieśnik, J. Electronic noses in classification and quality control of edible oils: A review (Review). Food Chem.
**2018**, 246, 192–201. [Google Scholar] [CrossRef] - Chen, H.Z.; Zhang, M.; Bhandari, B.; Guo, Z. Evaluation of the freshness of fresh-cut green bell pepper (Capsicum annuum var. grossum) using electronic nose. LWT-Food Sci. Tech.
**2018**, 87, 77–84. [Google Scholar] [CrossRef][Green Version] - Laura, B.; Francesca, C.; Serena, D.; Massimiliano, M.; Enrico, V.; Antonio, M.; Stefano, P. Sensory evaluation and instrumental measurements to determine tactile properties of wool fabrics. Text. Res. J.
**2012**, 14, 1430–1441. [Google Scholar] - Tiggemann, L.; Tiggemann, L.; Ballen, S.; Bocalon, C.; Graboski, A.M.; Manzoli, A.; de Paula, H.; Paulo, S.; Steffens, J.; Valduga, E.; et al. Low-cost gas sensors with polyaniline film for aroma detection. J. Food Eng.
**2016**, 180, 16–21. [Google Scholar] [CrossRef] - Men, H.; Shi, Y.; Fu, S.L.; Jiao, Y.N.; Qiao, Y.; Liu, J.J. Mining Feature of Data Fusion in the Classification of Beer Flavor Information Using E-Tongue and E-Nose. Sensors
**2017**, 7, 1656–1673. [Google Scholar] - Cetó, X.; González-Calabuig, A.; Crespo, N.; Pérez, S.; Capdevila, J.; Puig-Pujol, A.; Del Valle, M. Electronic tongues to assess wine sensory descriptors. Talanta
**2017**, 162, 218–224. [Google Scholar] - Xu, M.; Wang, J.; Zhu, L.Y. The qualitative and quantitative assessment of tea quality based on E-nose, E-tongue and E-eye combined with chemometrics. Food Chem.
**2019**, 289, 482–489. [Google Scholar] [CrossRef] - Wang, Z.C.; Yan, Y.Z.; Nisar, T.; Sun, L.J.; Zeng, Y.; Guo, Y.R.; Wang, H.C.; Fang, Z.X. Multivariate statistical analysis combined with e-nose and e-tongue assays simplifies the tracing of geographical origins of Lycium ruthenicum Murray grown in China. Food Contr.
**2019**, 98, 457–464. [Google Scholar] [CrossRef] - Zulkafli, M.; Tang, T. Electronic tongue for fresh milk assessment A revisit of using pH as indicator. IEEE Int. Conf. Circuits Syst.
**2013**, 9, 167–171. [Google Scholar] - Tian, X.J.; Wang, J.; Shen, R.Q.; Ma, Z.R.; Li, M.S. Discrimination of pork/chicken adulteration in minced mutton by electronic taste system. Int. J. Food Sci. Technol.
**2019**, 3, 670–678. [Google Scholar] [CrossRef] - Dai, C.X.; Huang, X.Y.; Huang, D.M.; Lv, R.Q.; Sun, J.; Zhang, Z.C.; Ma, M.; Aheto, J.H. Detection of submerged fermentation of Tremella aurantialba using data fusion of electronic nose and tongue. J. Food Process Eng.
**2019**, 87, 3. [Google Scholar] [CrossRef] - Wang, J.; Zhu, L.Y.; Zhang, W.L.; Wei, Z.B. Application of the voltammetric electronic tongue based on nanocomposite modified electrodes for identifying rice wines of different geographical origins. Anal. Chim. Acta
**2019**, 1050, 60–70. [Google Scholar] [CrossRef] - Milovanovic, M.; Žeravík, J.; Obořil, M.; Pelcová, M.; Lacina, K.; Cakar, U.; Petrovic, A.; Glatz, Z.; Skládal, P. A novel method for classification of wine based on organic acids. Food Chem.
**2019**, 284, 296–302. [Google Scholar] [CrossRef] - Banerjee, M.B.; Roy, R.B.; Tudu, B. Black tea classification employing feature fusion of E-Nose and E-Tongue responses. J. Food Eng.
**2019**, 224, 55–63. [Google Scholar] [CrossRef] - Zhang, L.; Wang, X.H.; Huang, G.B.; Liu, T.; Tan, X.H. Taste Recognition in E-Tongue Using Local Discriminant Preservation Projection. IEEE Trans. Cybern.
**2018**, 49, 947–960. [Google Scholar] [CrossRef] - Haddi, Z.; Mabrouk, S.; Bougrini, M.; Tahri, K.; Sghaier, K.; Barhoumi, H.; El Bari, N.; Maaref, A.; Jaffrezic-Renault, N.; Bouchikhi, B. E-Nose and e-Tongue combination for improved recognition of fruit juice samples. Food Chem.
**2014**, 150, 246–253. [Google Scholar] [CrossRef] [PubMed] - Dias, L.; Rodrigues, N.; Veloso, A.; Pereira, J.; Peres, A. Monovarietal extra-virgin olive oil classification: A fusion of human sensory attributes and an electronic tongue. Eur. Food Res. Technol.
**2016**, 2, 259–270. [Google Scholar] [CrossRef][Green Version] - Fekete, D.; Balazs, G.; Bohm, V.; Varvolgyi, E.; Kappel, N. Sensory Evaluation and Electronic Tongue for Sensing Grafted and Non-Grafted Watermelon Taste Attributes. Acta Aliment.
**2018**, 47, 487–494. [Google Scholar] [CrossRef] - Jeon, S.Y.; Kim, J.S.; Kim, G.C.; Choi, S.Y.; Kim, S.B.; Kim, K.M. Analysis of Electronic Nose and Electronic Tongue and Sensory Characteristics of Commercial Seasonings. Korean J. Food Cook. Sci.
**2017**, 33, 538–550. [Google Scholar] [CrossRef] - Kang, B.S.; Lee, J.E.; Park, H.J. Electronic tongue-based discrimination of Korean rice wines (makgeolli) including prediction of sensory evaluation and instrumental measurements. Food Chem.
**2014**, 151, 317–323. [Google Scholar] [CrossRef] - Liu, J.J.; Yang, J.L.; Wang, W.; Fu, S.L.; Shi, Y.; Men, H. Automatic Evaluation of Sensory Information for Beer at a Fuzzy Level Using Electronic Tongue and Electronic Nose. Sens. Mater.
**2016**, 7, 785–795. [Google Scholar] - Zhang, J.N.; Zhao, M.M.; Su, G.W.; Lin, L.Z. Identification and taste characteristics of novel umami and umami-enhancing peptides separated from peanut protein isolate hydrolysate by consecutive chromatography and UPLC-ESI-QTOF-MS/MS. Food Chem.
**2019**, 278, 674–682. [Google Scholar] [CrossRef] - Barnett, S.M.; Diako, C.; Ross, C.F. Identification of a Salt Blend: Application of the Electronic Tongue, Consumer Evaluation, and Mixture Design Methodology. J. Food Sci.
**2019**, 2, 327–338. [Google Scholar] [CrossRef] - Yu, H.Y.; Zhang, Y.; Zhao, J.; Tian, H.X. Taste characteristics of Chinese bayberry juice characterized by sensory evaluation, chromatography analysis, and an electronic tongue. J. Food Sci. Technol.-Mysore
**2018**, 5, 1624–1631. [Google Scholar] [CrossRef] [PubMed] - Sipos, L.; Attila, G.; Szöllősi, D.; Kovács, Z.; Kókai, Z.; Fekete, A. Sensory evaluation and electronic tongue for sensing flavored mineral water taste attributes. J. Food Sci.
**2013**, 10, S1602–S1608. [Google Scholar] [CrossRef] [PubMed] - Xu, Q.W.; Xu, K.L. Quality evaluation of Chinese red wine based on cloud model. J. Food Biochem.
**2019**, 43, 1–11. [Google Scholar] [CrossRef] [PubMed] - China National Standardization Administration Committee. GB/T 33405-2016 Liquor Sensory Products Evaluation Guide; China Standard Press: Beijing, China, 2016. [Google Scholar]
- Shi, Y.; Gong, F.R.; Wang, M.; Liu, J.J.; Wu, Y.; Men, H. A deep feature mining method of electronic nose sensor data for identifying beer olfactory information. J. Food Eng.
**2019**, 263, 437–445. [Google Scholar] [CrossRef] - Li, D.Y.; Meng, H.J.; Shi, X.M. Affiliated cloud and affiliated cloud generator. J. Comput. Res. Dev.
**1995**, 32, 15–20. [Google Scholar] - Liu, C.Y.; Li, D.Y.; Pan, L.L. Uncertainty knowledge representation based on cloud model. Comput. Eng. Appl.
**2004**, 2, 32–35. [Google Scholar]

**Figure 1.**The SA-402B e-tongue system (a is used to measure the aftertaste value, b–c is used to quickly clean the sample, d–e is used to clean the positive and negative solution, f is the positive and negative cleaning solution, g is applied for sensor calibration, h is used for sensor reset, and i is liquor sample).

**Figure 2.**Classification process for liquor taste information with GA-SVM: (

**a**) The parameter optimization fitness curve. (

**b**) The classification result.

**Figure 3.**Cloud drop results of four different flavor types of liquor (●: jiang-flavor style, ●: feng-flavor style, ●: nong-flavor style, ●: mild-flavor style).

**Figure 4.**The proportion of liquor taste words selected by examiners: (

**a**) jiang-flavor style, (

**b**) feng-flavor style, (

**c**) nong-flavor style, and (

**d**) mild-flavor style.

**Figure 5.**Liquor taste information cloud model division result (purple: jiang-flavor style; blue: feng-flavor style; green: nong-flavor style; brown: mild-flavor style).

**Figure 6.**Cloud drop test of 8 groups of liquor data: (

**a**) jiang-flavor style-1; (

**b**) jiang-flavor style-2; (

**c**) feng-flavor style-1; (

**d**) feng-flavor style-2; (

**e**) nong-flavor style-1; (

**f**) nong-flavor style-2; (

**g**) mild-flavor style-1; (

**h**) mild-flavor style-2.

Liquor Name | Flavor | Raw Material | Alcohol Content (% vol) | Manufacturer |
---|---|---|---|---|

Sauce incense private 1979 | jiang | Water, sorghum, wheat | 53 | Shijia Wine Industry Co., Ltd. |

Xifeng wine | feng | Water, sorghum, barley, wheat, peas | 55 | Shanxi Xifeng Wine Co., Ltd. |

Sealed puree wine V60 | nong | Water, sorghum, wheat, rice, corn, glutinous rice | 52 | Ziyunting Wine Co., Ltd. |

Red Star Erguotou | mild | Sorghum, water, corn, barley, peas | 52 | Beijing Red Star Co., Ltd. |

Flavor | Liquor Taste Description Words |
---|---|

jiang | Elegant and delicate, Fully mellow, Full bodied, Long aftertaste, Coordination |

feng | Mellow fullness, Sweet and cool, Mellow and elegant, All tastes harmonize, Long clean tail |

nong | Alcohol harmonious, Sweet and refreshing, Soft and sweet, Long aftertaste, Mellow |

mild | Pure fragrance, Sweet and soft, Natural coordination, Sweet and refreshing, Long aftertaste |

Flavor | Ex | En | He | |||
---|---|---|---|---|---|---|

PC1 | PC2 | PC1 | PC2 | PC1 | PC2 | |

jiang | 41.6501 | −23.6840 | 2.7506 | 2.3288 | 1.2743 | 1.1007 |

feng | −31.8733 | −23.6339 | 2.6604 | 3.6078 | 0.3852 | 1.8173 |

nong | 30.2356 | −27.0657 | 2.9954 | 2.3639 | 1.3505 | 1.6552 |

mild | 4.4596 | −36.4820 | 3.3641 | 2.1433 | 0.5800 | 0.7641 |

Serial Number | The Actual Flavor of Liquor | Predicted Flavor of Liquor | Cloud Droplets Areas | Evaluation Language |
---|---|---|---|---|

a | jiang | jiang | $\frac{{(x-41.6501)}^{2}}{{2.8881}^{2}}+\frac{{(y+23.684)}^{2}}{{2.4452}^{2}}\le 1$; $\begin{array}{l}\frac{{(x-41.6501)}^{2}}{{4.1259}^{2}}+\frac{{(y+23.684)}^{2}}{{3.4932}^{2}}\le 1\\ \&\frac{{(x-41.6501)}^{2}}{{2.8881}^{2}}+\frac{{(y+23.684)}^{2}}{{2.4452}^{2}}>1\end{array}$; $\begin{array}{l}\frac{{(x-41.6501)}^{2}}{{8.2518}^{2}}+\frac{{(y+23.684)}^{2}}{{6.9864}^{2}}\le 1\\ \&\frac{{(x-41.6501)}^{2}}{{6.1613}^{2}}+\frac{{(y+23.684)}^{2}}{{5.2165}^{2}}>1\end{array}$ | This liquor is fully mellow, elegant and delicate, coordination |

b | jiang | jiang | $\frac{{(x-41.6501)}^{2}}{{2.8881}^{2}}+\frac{{(y+23.684)}^{2}}{{2.4452}^{2}}\le 1$; $\begin{array}{l}\frac{{(x-41.6501)}^{2}}{{4.84}^{2}}+\frac{{(y+23.684)}^{2}}{{4.0987}^{2}}\le 1\\ \&\frac{{(x-41.6501)}^{2}}{{4.1259}^{2}}+\frac{{(y+23.684)}^{2}}{{3.4932}^{2}}>1\end{array}$; $\begin{array}{l}\frac{{(x-41.6501)}^{2}}{{6.1613}^{2}}+\frac{{(y+23.684)}^{2}}{{5.2165}^{2}}\le 1\\ \&\frac{{(x-41.6501)}^{2}}{{4.84}^{2}}+\frac{{(y+23.684)}^{2}}{{4.0987}^{2}}>1\end{array}$ | This liquor is fully mellow, full bodied, long aftertaste |

c | feng | feng | $\frac{{(x+31.8733)}^{2}}{{2.2081}^{2}}+\frac{{(y+23.6339)}^{2}}{{2.9945}^{2}}\le 1$; $\begin{array}{l}\frac{{(x+31.8733)}^{2}}{{4.6823}^{2}}+\frac{{(y+23.6339)}^{2}}{{6.3497}^{2}}\le 1\\ \&\frac{{(x+31.8733)}^{2}}{{3.4585}^{2}}+\frac{{(y+23.6339)}^{2}}{{4.6901}^{2}}>1\end{array}$ | This liquor is sweet and cool, mellow and elegant |

d | feng | feng | $\frac{{(x+31.8733)}^{2}}{{2.2081}^{2}}+\frac{{(y+23.6339)}^{2}}{{2.9945}^{2}}\le 1$; $\begin{array}{l}\frac{{(x+31.8733)}^{2}}{{3.4585}^{2}}+\frac{{(y+23.6339)}^{2}}{{4.6901}^{2}}\le 1\\ \&\frac{{(x+31.8733)}^{2}}{{2.2081}^{2}}+\frac{{(y+23.6339)}^{2}}{{2.9945}^{2}}>1\end{array}$; $\begin{array}{l}\frac{{(x+31.8733)}^{2}}{{4.6823}^{2}}+\frac{{(y+23.6339)}^{2}}{{6.3497}^{2}}\le 1\\ \&\frac{{(x+31.8733)}^{2}}{{3.4585}^{2}}+\frac{{(y+23.6339)}^{2}}{{4.6901}^{2}}>1\end{array}$ $\begin{array}{l}\frac{{(x+31.8733)}^{2}}{{5.9593}^{2}}+\frac{{(y+23.6339)}^{2}}{{8.0815}^{2}}\le 1\\ \&\frac{{(x+31.8733)}^{2}}{{4.6823}^{2}}+\frac{{(y+23.6339)}^{2}}{{6.3497}^{2}}>1\end{array}$ | This liquor is sweet and cool, long clean tail, mellow and elegant, all tastes harmonize |

e | nong | nong | $\frac{{(x-30.2356)}^{2}}{{2.6959}^{2}}+\frac{{(y+27.0657)}^{2}}{{2.1275}^{2}}\le 1$; $\begin{array}{l}\frac{{(x-30.2356)}^{2}}{{4.1636}^{2}}+\frac{{(y+27.0657)}^{2}}{{3.2858}^{2}}\le 1\\ \&\frac{{(x-30.2356)}^{2}}{{2.6959}^{2}}+\frac{{(y+27.0657)}^{2}}{{2.1275}^{2}}>1\end{array}$; $\begin{array}{l}\frac{{(x-30.2356)}^{2}}{{5.8410}^{2}}+\frac{{(y+27.0657)}^{2}}{{4.6096}^{2}}\le 1\\ \&\frac{{(x-30.2356)}^{2}}{{4.1636}^{2}}+\frac{{(y+27.0657)}^{2}}{{3.2858}^{2}}>1\end{array}$; $\begin{array}{l}\frac{{(x-30.2356)}^{2}}{{8.9862}^{2}}+\frac{{(y+27.0657)}^{2}}{{7.0917}^{2}}\le 1\\ \&\frac{{(x-30.2356)}^{2}}{{6.7097}^{2}}+\frac{{(y+27.0657)}^{2}}{{5.2951}^{2}}>1\end{array}$ | This liquor is soft and sweet, sweet and refreshing, Mellow, long aftertaste |

f | nong | nong | $\frac{{(x-30.2356)}^{2}}{{2.6959}^{2}}+\frac{{(y+27.0657)}^{2}}{{2.1275}^{2}}\le 1$; $\begin{array}{l}\frac{{(x-30.2356)}^{2}}{{4.1636}^{2}}+\frac{{(y+27.0657)}^{2}}{{3.2858}^{2}}\le 1\\ \&\frac{{(x-30.2356)}^{2}}{{2.6959}^{2}}+\frac{{(y+27.0657)}^{2}}{{2.1275}^{2}}>1\end{array}$; $\begin{array}{l}\frac{{(x-30.2356)}^{2}}{{5.8410}^{2}}+\frac{{(y+27.0657)}^{2}}{{4.6096}^{2}}\le 1\\ \&\frac{{(x-30.2356)}^{2}}{{4.1636}^{2}}+\frac{{(y+27.0657)}^{2}}{{3.2858}^{2}}>1\end{array}$ | This liquor is soft and sweet, sweet and refreshing, mellow |

g | mild | mild | $\frac{{(x-4.4596)}^{2}}{{2.523}^{2}}+\frac{{(y+36.482)}^{2}}{{1.6075}^{2}}\le 1$; $\begin{array}{l}\frac{{(x-4.4596)}^{2}}{{4.0706}^{2}}+\frac{{(y+36.482)}^{2}}{{2.5934}^{2}}\le 1\\ \&\frac{{(x-4.4596)}^{2}}{{2.523}^{2}}+\frac{{(y+36.482)}^{2}}{{1.6075}^{2}}>1\end{array}$; $\begin{array}{l}\frac{{(x-4.4596)}^{2}}{{10.0923}^{2}}+\frac{{(y+36.482)}^{2}}{{6.4299}^{2}}\le 1\\ \&\frac{{(x-4.4596)}^{2}}{{6.56}^{2}}+\frac{{(y+36.482)}^{2}}{{4.1794}^{2}}>1\end{array}$ | This liquor is pure fragrance, long aftertaste, sweet and refreshing |

h | mild | mild | $\begin{array}{l}\frac{{(x-4.4596)}^{2}}{{4.0706}^{2}}+\frac{{(y+36.482)}^{2}}{{2.5934}^{2}}\le 1\\ \&\frac{{(x-4.4596)}^{2}}{{2.523}^{2}}+\frac{{(y+36.482)}^{2}}{{1.6075}^{2}}>1\end{array}$; $\begin{array}{l}\frac{{(x-4.4596)}^{2}}{{5.0462}^{2}}+\frac{{(y+36.482)}^{2}}{{3.2150}^{2}}\le 1\\ \&\frac{{(x-4.4596)}^{2}}{{4.0706}^{2}}+\frac{{(y+36.482)}^{2}}{{2.5934}^{2}}>1\end{array}$; $\begin{array}{l}\frac{{(x-4.4596)}^{2}}{{6.56}^{2}}+\frac{{(y+36.482)}^{2}}{{4.1794}^{2}}\le 1\\ \&\frac{{(x-4.4596)}^{2}}{{5.0462}^{2}}+\frac{{(y+36.482)}^{2}}{{3.2150}^{2}}>1\end{array}$; $\begin{array}{l}\frac{{(x-4.4596)}^{2}}{{10.0923}^{2}}+\frac{{(y+36.482)}^{2}}{{6.4299}^{2}}\le 1\\ \&\frac{{(x-4.4596)}^{2}}{{6.56}^{2}}+\frac{{(y+36.482)}^{2}}{{4.1794}^{2}}>1\end{array}$ | This liquor is long aftertaste, sweet and soft, natural coordination, sweet and refreshing |

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Liu, J.; Zuo, M.; Low, S.S.; Xu, N.; Chen, Z.; Lv, C.; Cui, Y.; Shi, Y.; Men, H. Fuzzy Evaluation Output of Taste Information for Liquor Using Electronic Tongue Based on Cloud Model. *Sensors* **2020**, *20*, 686.
https://doi.org/10.3390/s20030686

**AMA Style**

Liu J, Zuo M, Low SS, Xu N, Chen Z, Lv C, Cui Y, Shi Y, Men H. Fuzzy Evaluation Output of Taste Information for Liquor Using Electronic Tongue Based on Cloud Model. *Sensors*. 2020; 20(3):686.
https://doi.org/10.3390/s20030686

**Chicago/Turabian Style**

Liu, Jingjing, Mingxu Zuo, Sze Shin Low, Ning Xu, Zhiqing Chen, Chuang Lv, Ying Cui, Yan Shi, and Hong Men. 2020. "Fuzzy Evaluation Output of Taste Information for Liquor Using Electronic Tongue Based on Cloud Model" *Sensors* 20, no. 3: 686.
https://doi.org/10.3390/s20030686