Momordica grosvenori variety quality typing accurate identification method and model

Abstract

The invention discloses a siraitia grosvenorii variety quality typing accurate identification method and model, and belongs to the technical field of medicinal material detection.The siraitia grosvenorii variety quality typing accurate identification method comprises the steps that 1, siraitia grosvenorii saponin quality indexes of a plurality of varieties are detected; the method comprises the following steps of: acquiring four main key content indexes, namely 11-O-mogroside V (11-O-MV), mogroside V (MV), isomogroside V (IMV) and siamenoside I (SI) in the momordica grosvenori, as well as a total saponin content index and a sweetness data index; step 2, carrying out clustering processing on the obtained mogroside content and sweetness data to generate a plurality of clustering categories; step 3, establishing a corresponding relationship between the quality of the siraitia grosvenorii variety and the siraitia grosvenorii saponin content and sweetness clustering category; and step 4, based on the established corresponding relationship, performing variety quality typing identification on the to-be-detected siraitia grosvenorii sample. According to the method, the efficiency and the precision degree of the siraitia grosvenorii variety quality typing identification can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medicinal material detection, and more specifically, to a method and model for accurately identifying the quality of monk fruit varieties by typing. Background Art

[0002] Momordica grosvenori Siraitia grosvenorii The fruit of the (Swingle) C.Jeffrey ex AM Lu et ZY Zhang is one of the first batch of precious Chinese medicinal materials for both medicine and food announced by the Ministry of Health. It is distributed in Guangxi, Guangdong, Guizhou, Hunan and Jiangxi provinces. Ancient herbal books and the Pharmacopoeia of the People's Republic of China record that monk fruit has the effects of clearing heat and moistening the lungs, relieving cough and asthma, relieving sore throat and opening voice, and moistening the intestines and relieving constipation. In addition to being used as a decoction piece, there are more than 90 Chinese patent medicines using monk fruit as raw materials, including 6 major dosage forms such as granules, tablets, capsules, tinctures, powders, and syrups. The main active ingredient of monk fruit, monk fruit saponin, is a natural substance with zero calories and high sweetness. It is a precious raw material for beverages, candies and other food industries, and an ideal substitute for sucrose. In recent years, the products using monk fruit saponin in the food and beverage industry at home and abroad, including the Coca-Cola Company of the United States, have increased year by year, and the number of products using monk fruit saponin sweeteners has exceeded 5,000.

[0003] Momordica fruit is rich in various glycosides. The sweetness of 11-O-mogroside V (11-O-MV), mogroside V (MV), isomogroside V (IMV) and simanoside I (SI) are 68 times, 378 times, 500 times and 465 times that of 0.5% sucrose solution, respectively. In addition to mogroside V which accounts for 30-45% of the total glycosides, the proportion of 11-O-mogroside V, isomogroside V, simanoside I and other glycosides in the total glycosides also reaches 20%. Different glycosides have medicinal activity. For example, mogroside V has anti-inflammatory effects, 11-O-mogroside V has antioxidant effects, and simanoside I has blood sugar and lipid-lowering effects. There are significant differences in the content of glycosides in different varieties. The current single indicator method for detecting the content of momordica saponin V is unable to comprehensively and accurately identify the varieties of momordica and its product quality. It is necessary to use the content index of multiple glycosides to identify the quality of momordica varieties. The detection and identification method of multiple glycosides alone will increase the difficulty of detection and identification. Therefore, it is urgent to design a method for typing and identifying the quality of momordica varieties to improve the efficiency and accuracy of identification. Summary of the invention

[0004] The invention aims to provide a method for accurately identifying the quality of Momordica grosvenori varieties by typing, which is used to quickly identify the quality of Momordica grosvenori varieties and improve the efficiency and accuracy of identification.

[0005] In order to achieve these objectives of the present invention, the present invention provides a method for accurately identifying the quality of Momordica grosvenori varieties by typing, comprising: a) Testing steps: Test several varieties of monk fruit to obtain the content data of monk fruit saponins and sweetness data indicators. The saponin content data includes at least the following four key content indicators: 11-O-monogroside (11-O-MV), mogroside V (MV), isomogroside V (IMV), simanoside I (SI) and the total saponin content indicator; b) Clustering processing step: clustering the acquired data on the content of mogrosides and the sweetness data to generate a number of cluster categories; c) Correspondence establishment step: establishing the correspondence between the quality of monk fruit varieties and the clustering categories of monk fruit saponin content and sweetness; d) Variety quality typing step: Based on the established correspondence, the variety quality typing of the monk fruit samples to be tested is carried out.

[0006] Preferably, the detecting step comprises: Sample collection: Collect monk fruit samples from multiple varieties to ensure that the samples are representative and clearly mark the variety information; Saponin extraction and detection: Extraction and quantitative analysis of mogroside saponins in each sample; Data recording: Record the content data of each saponin detected, and calculate the total saponin content and sweetness data indicators to form a complete set of monk fruit saponin content and sweetness data.

[0007] Preferably, total saponin content = sum of SI content + IMV content + MV content + 11-O-MV content; sweetness index is: sweetness relative to sucrose = (SI content × 465 + IMV content × 500 + MV content × 378 + 11-O-MV content × 68) / 5.

[0008] Preferably, the saponin components detected in the saponin extraction and detection step may also include any one or more of mogroside IIE (MIIE), mogroside III (MIII), mogroside IV (MIV), mogroside IVa (MIVa) and the like.

[0009] Preferably, the clustering processing step includes: Data preprocessing: The data of mogroside content and sweetness were processed using standardized methods; Select clustering algorithm: Select a suitable clustering algorithm based on the characteristics of the data and clustering requirements; Set clustering parameters: set the number of clusters K, metric, and minimum number of samples; Perform clustering processing: input the pre-processed momordica saponin content and sweetness data into the clustering algorithm, perform clustering processing, and generate a number of cluster categories.

[0010] Preferably, the clustering algorithm selected in the step of selecting the clustering algorithm is the K-means clustering algorithm.

[0011] Preferably, in the step of establishing the corresponding relationship, according to the clustering result, a corresponding relationship is established between the variety quality of Siraitia grosvenorii and the corresponding clustering category, forming a corresponding relationship table between variety quality and category.

[0012] Preferably, in the step of classifying the variety quality, the saponin content and sweetness data of the Siraitia grosvenorii sample to be tested are compared with the established corresponding relationship to determine the variety quality of the Siraitia grosvenorii sample to be tested.

[0013] A precise identification model for classifying the variety quality of Siraitia grosvenorii, which is constructed based on the precise identification method for classifying the variety quality of Siraitia grosvenorii described in any one of the above, and is used to classify the variety and product quality of Siraitia grosvenorii samples.

[0014] Preferably, the model outputs all possible variety quality classification results and sends them to the user, and the user determines the variety quality according to the results.

[0015] Preferably, the accuracy of the quality classification identification of the model can be improved with the increase in the types of saponins detected.

[0016] Specifically, due to the complex and diverse components of Siraitia grosvenorii and the weak ultraviolet absorption of Siraitia grosvenorii saponin components, the determination by high performance liquid chromatography is severely interfered by other strong ultraviolet absorption compounds, and has high requirements for chromatographic separation, and the detection method has low sensitivity. Therefore, the present invention provides a method for detecting the saponin content of Siraitia grosvenorii for the above analysis method, which uses liquid chromatography-tandem mass spectrometry, and has the advantages of less solvent consumption, high sensitivity, good selectivity, etc., and can simultaneously detect at least 8 kinds of Siraitia grosvenorii saponins. This method includes: Step 1, prepare a standard stock solution of Siraitia grosvenorii saponin compounds. Accurately weigh the standards of several substances to be tested, dissolve them with methanol, and quantitatively dilute them with 80% methanol solution to prepare a high-concentration single standard stock solution, and store it as a mother liquor of 0.1 mg / mL. Step 2, prepare a mixed standard solution of Siraitia grosvenorii saponin compounds. Dilute the mother liquor with 80% methanol solution to prepare mixed standard solutions with different concentrations for drawing a standard curve. Step 3, sample preparation. Mix the Siraitia grosvenorii samples evenly, take out a part as the sample, crush it and mix it evenly through a 20-mesh sieve and seal it.

[0017] Step 4, sample extraction. Weigh 0.5 g of Siraitia grosvenorii powder, add 25 mL of 80% methanol solution, ultrasonically extract it at room temperature for 30 minutes, centrifuge it, and combine the supernatant and make the volume up to 100 mL.

[0018] Step 5: Perform detection by liquid chromatography-tandem mass spectrometry. The ultra-high performance liquid chromatography conditions for the detection include using an Agilent Poroshell 120 SB C18 chromatographic column, with mobile phase A being an aqueous solution of 0.1% formic acid, mobile phase B being acetonitrile, a flow rate of 0.2 mL / min, an injection volume of 2 μL, and gradient elution being performed. Set the mass spectrometry conditions as negative ion scanning, with an ionization voltage of 4500 V, an ion source temperature of 550 °C, and detection being performed using the multiple reaction monitoring (MRM) method.

[0019] Step 6: Perform the determination of the test sample to obtain the total ion chromatogram and the ion chromatograms of each component of the mixed standard solution and the Siraitia grosvenorii sample.

[0020] Step 7: Perform qualitative measurement. Compare the retention time of the target chromatographic peak in the sample solution and the standard solution and the relative abundance of the mass spectrometry ion pairs to determine whether the component exists in the sample.

[0021] Step 8: Perform quantitative measurement. Use the concentration of the mixed standard solution at different concentrations as the abscissa and the chromatographic peak area as the ordinate to plot a standard curve, with a correlation coefficient not less than 0.99, ensuring that the response values of the sample solution and the standard solution are within the linear range of the instrument detection.

[0022] Step 9: Calculate the results. Use the formula to calculate the content of Siraitia grosvenorii saponins in the test sample.

[0023] The present invention at least includes the following beneficial effects: 1. The method and model for accurate identification of the variety quality typing of Siraitia grosvenorii provided by the present invention, by detecting the content of the main key saponins in Siraitia grosvenorii, combined with clustering processing and establishing corresponding relationships, realizes the rapid and accurate typing of the variety quality of Siraitia grosvenorii, as well as the discrimination of the saponin content and sweetness of the variety. This method has the advantages of simple operation and reliable results.

[0024] 2. Due to the complex and diverse components of Siraitia grosvenorii, and the weak ultraviolet absorption of Siraitia grosvenorii saponin components, the traditional determination using high performance liquid chromatography is severely interfered by other compounds with strong ultraviolet absorption, and has high requirements for chromatographic separation, with low detection method sensitivity. However, the method for detecting the content of Siraitia grosvenorii saponins of the present invention uses liquid chromatography-tandem mass spectrometry, which has the advantages of less solvent consumption, high sensitivity, and good selectivity.

[0025] Other advantages, objectives, and features of the present invention will be partially reflected by the following description, and partially will also be understood by those skilled in the art through the research and practice of the present invention. Description of the Drawings

[0026] Figure 1Total ion chromatogram of 8 mogrosides in the mixed standard solution of Example 1; Figure 2 Ion chromatograms of each component in Example 1; Figure 3 Data in Table 5. Detailed implementation mode

[0027] The present invention will be further described in detail below in conjunction with examples, so that those skilled in the art can implement it according to the text of the specification.

[0028] It should be understood that terms such as "having", "comprising" and "including" used herein do not exclude the presence or addition of one or more other elements or their combinations.

[0029] The mogroside compound reference standards used in the present invention: the relevant information is shown in Table 1, and the purity is not less than 98%.

[0030] Table 1 Mogroside compound reference standards Example 1 This example gives a method example for simultaneously determining the contents of 8 glucosides in Siraitia grosvenorii varieties and their products by liquid chromatography-tandem mass spectrometry, including: Step 1, mogroside compound standard stock solution: accurately weigh the reference standards of 8 substances to be measured, dissolve them with methanol, and quantitatively dilute them with 80% methanol solution to prepare high-concentration single reference standard stock solutions; respectively transfer a certain amount of 8 single reference standard stock solutions into a 25 mL volumetric flask, and make up the volume with 80% methanol solution to prepare a mother liquor of 0.1 mg / mL as the reference standard stock solution, and store it in the dark at 2-4 °C for later use.

[0031] Step 2, mogroside compound mixed standard solution: dilute the mother liquor of 0.1 mg / mL with 80% methanol solution to prepare mixed standard solutions with different concentrations for drawing the standard curve.

[0032] Step 3, sample preparation: mix the Siraitia grosvenorii samples evenly, take out 0.5 kg as the sample, crush it with a pulverizer and pass through a 20-mesh sieve, mix well and seal it, and make a mark.

[0033] Step 4, Sample extraction: Weigh 0.5 g of Momordica grosvenori powder, place it in a 50 mL centrifuge tube, add 25 mL of 80% methanol solution, and ultrasonically extract for 30 min at room temperature; centrifuge at 7500 r / min for 5 min, transfer the supernatant to a 100 mL volumetric flask, repeat the extraction of the precipitate with 80% methanol solution once, centrifuge for 5 min under the same conditions, combine the supernatants, and make up to 100 mL with 80% methanol solution; before loading the sample, filter it through a 0.22 μm organic phase filter membrane.

[0034] Step 5, Determination by liquid chromatography - tandem mass spectrometry: Ultra - performance liquid chromatography reference conditions Chromatographic column: Agilent Poroshell 120 SB C18 (100 mm × 2.1 mm, 2.7 μm); Mobile phase A: 0.1% formic acid aqueous solution, take 0.1 mL of formic acid and make up to 100 mL with water, mix well; Mobile phase B: Acetonitrile; Column temperature: Room temperature; Flow rate: 0.2 mL / min; Injection volume: 2 μL; Mobile phase: The gradient elution program is as follows in the table; Table 2 Mobile phase and gradient elution program Mass spectrometry conditions Scanning mode: Negative ion scanning; Ionization voltage: 4500 V; Ion source temperature: 550 °C; Curtain gas: Nitrogen (purity above 97%), pressure 20 psi; Nebulizing gas: Nitrogen (purity above 97%), gas pressure 55 psi; Auxiliary gas: Nitrogen (purity above 97%), gas pressure 45 psi; Detection mode: Multiple reaction monitoring (MRM); The retention times and monitored ions of Momordica grosvenori saponins are as follows in the table: Table 3 Information on the retention times and monitored ion pairs of Momordica grosvenori saponins Note: q is the quantitative ion.

[0035] Step 6. Determination of the test sample: Under the optimal conditions of the instrument, take the mixed standard solution of mogroside compounds and the momordica grosvenori sample for on-machine determination respectively, and obtain the total ion chromatogram of 8 mogrosides in the mixed standard solution ( Figure 1 ), and the ion chromatogram of each component ( Figure 2 ).

[0036] Step 7. Qualitative measurement: Measure the test sample and the mixed standard solution under the same test conditions. If the retention time of the chromatographic peak detected in the sample solution is consistent with that of the target chromatographic peak in the matrix standard solution (the variation range is ±2.5%), and the relative abundance of the mass spectrometry ion pairs in the sample solution is compared with that of the matrix standard solution, and the relative deviation of the relative ion abundance does not exceed the range specified in the following table, it can be determined that the component exists in the sample.

[0037] Table 4 Maximum allowable error of relative ion abundance during qualitative determination Step 8. Quantitative measurement Taking the concentration of the mixed standard solution at different concentrations as the abscissa and the chromatographic peak area (response value) as the ordinate, draw a standard curve. The standard curve in this example is: Y = 69904.9 + 660708*X, and the correlation coefficient R 2 = 0.9966; the correlation coefficient of the standard curve should not be lower than 0.99; the response values of the analytes in the test sample solution and the standard solution should both be within the linear range detected by the instrument; if it exceeds the linear range, the test should be repeated or the test sample solution and the matrix-matched mixed standard series solution should be diluted accordingly and then re-determined; when calibrating quantitatively with a single point, the concentration of the analyte in the test sample solution should not differ from the concentration of the solvent mixed standard solution by more than 30%.

[0038] Step 9. Result calculation Use liquid chromatography-mass spectrometry data processing software or calculate according to the formula the amount of the detection target in the test sample, and calculate according to formula (1): (1) In the formula: W—the value of the content of mogroside compounds in the test sample, unit: milligram per kilogram (mg / kg); A—the peak area of mogroside compounds in the sample; ρi—the value of the standard solution concentration, unit: milligram per liter (mg / L); V—the value of the constant volume, unit: milliliter (mL); f—the dilution factor; Ai—the peak area of mogroside compounds in the standard sample; m—the value of the mass of the sample represented by the final solution, unit: gram (g).

[0039] Note: The blank value should be deducted from the calculation result. The determination result is expressed as the arithmetic mean of parallel determinations, and three significant figures are retained.

[0040] Example 2 This example gives an example of a precise identification method and model for the quality typing of Momordica grosvenori varieties Specifically, it includes the following steps: Step 1: Detect the quality of Momordica grosvenori saponins of several varieties according to the method of Example 1 to obtain the data of Momordica grosvenori saponin content 1) Sample collection: Collect Momordica grosvenori samples from multiple varieties in several production areas to ensure the representativeness of the samples. Each sample should include a complete Momordica grosvenori fruit, and the variety information should be clearly marked.

[0041] 2) Saponin extraction and detection: Adopt the method of Example 1 to extract and quantitatively analyze Momordica grosvenori saponins in each sample. The saponin components to be detected include at least four main key content indicators: 11-O-Momordica grosvenori saponin (11-O-MV), Momordica grosvenori saponin V (MV), isomomordica grosvenori saponin V (IMV), and semenoside I (SI). The total saponin content (the sum of these four key content indicators) is counted as an evaluation dimension, and the sweetness relative to sucrose (the sum of SI content × 465 + IMV content × 500 + MV content × 378 + 11-O-MV content × 68 divided by 5) is counted as an evaluation dimension.

[0042] Optionally, any one or more of the indicators such as Momordica grosvenori saponin IIE (MIIE), Momordica grosvenori saponin III (MIII), Momordica grosvenori saponin IV (MIV), and Momordica grosvenori saponin IVa (MIVa) can also be considered.

[0043] The more components compared, the more dimensions of comparison, and the more accurate the result. However, some indicators do not show significant differences among various varieties. Therefore, selecting the above main key indicators can ensure accuracy while improving efficiency.

[0044] 3) Data recording: Record the content data of each saponin obtained from the detection, and calculate the total glycoside content and the sweetness multiple relative to sucrose to form a complete dataset of Momordica grosvenori saponin content and sweetness. In this example, at least 3 sample content data are made for each variety. Total saponin content = the sum of SI content + IMV content + MV content + 11-O-MV content; Sweetness relative to sucrose = (SI content × 465 + IMV content × 500 + MV content × 378 + 11-O-MV content × 68) / 5 The dataset is as shown in the following table or Figure 3 as follows: Table 5 Mogroside content and sweetness data of each variety Step 2: Construct a precise identification model for the quality classification of Momordica grosvenori varieties, perform clustering on the mogroside content and sweetness data, and generate several clustering categories In this embodiment, a clustering algorithm model is used, which specifically includes: 1) Data preprocessing: For the above-mentioned mogroside content data and sweetness data, use the Z-score normalization method for processing and perform normalization processing.

[0045] 2) Select a clustering algorithm: According to the characteristics of the data and clustering requirements, select a suitable clustering algorithm, such as K-means clustering, hierarchical clustering, or DBSCAN clustering, etc. In this embodiment, the K-means clustering algorithm is selected.

[0046] 3) Set clustering parameters: Set the number of clusters K = 2, select the Euclidean distance as the metric standard, and set the minimum number of samples to 4. Among them, both K and the minimum number of samples can be set according to actual needs and are not limited to the above settings.

[0047] 4) Perform clustering processing: Input the preprocessed mogroside content and sweetness data into the clustering algorithm, perform clustering processing, and generate several clustering categories. In this embodiment, K = 2, and 2 clustering categories are obtained. It is also possible to set more than 2 clustering categories according to actual needs.

[0048] The clustering results are as follows in the table: Table 6 Clustering results of each saponin content Table 7 Clustering results of total saponin content Table 8 Clustering results relative to the sweetness of sucrose It can be seen that the content difference of mogroside IIE (MIIE) among varieties is not significant, and it can be selected for elimination. MIII can be selected for consideration, but in this embodiment, the above four main key content indicators, as well as the total saponin content and sweetness dimension, are mainly selected as examples.

[0049] Step 3: Establish the correspondence between the quality of Momordica grosvenori varieties and the clustering categories of mogroside content For example, for the clustering results of the above-mentioned 11-O-MV content, MV content, IMV content, SI content, total saponin content, and sweetness content, establish the following correspondence: Table 9 Correspondence between saponin content and categories The following is a complete corresponding relationship table of varieties and categories obtained: Table 10 Corresponding Relationship between Varieties and Categories According to the above corresponding relationship table, the mean values of samples of the same variety (at least 3 samples of the same variety were tested in this embodiment) are compared to determine the category of the quality of this variety. For example, the category of variety 174 is: low low low high medium medium (in order: 11-O-mogroside (11-O-MV), mogroside V (MV), isomogroside V (IMV), semenicoside I (SI), total saponin content, sweetness). When applying, the mogroside content and sweetness data of the to-be-detected Momordica grosvenori samples are compared with the relationship table, and the quality type of the Momordica grosvenori variety can be quickly judged.

[0050] It should be noted that the above corresponding relationship table only gives an example. As the types of mogrosides (with significant differences in variety content) increase, the dimensions of comparison increase, and the results obtained are more accurate.

[0051] Step 4: Based on the above-established corresponding relationship, the mogroside content and sweetness data detected from the to-be-detected Momordica grosvenori samples are input into the precise identification model for the quality typing of Momordica grosvenori varieties, and then the quality typing of the variety can be carried out, and the heights in terms of the total saponin content and sweetness can be understood. The result of the quality typing of the variety obtained is sent to the user, and the user can quickly determine the variety quality or the variety quality range according to the result.

[0052] The precise identification method for the quality typing of Momordica grosvenori varieties proposed in this embodiment realizes the rapid and accurate typing of the quality of Momordica grosvenori varieties by detecting the content of the main key saponins in Momordica grosvenori, combining clustering processing and establishing corresponding relationships. This method has the advantages of simple operation and reliable results, and provides strong support for the precise identification of the quality of Momordica grosvenori variety resources and the quality control, development and utilization of its medicinal materials.

[0053] Although the implementation scheme of the present invention is disclosed as above, it is not limited to only the applications listed in the description and the implementation mode. It can be fully applied to various fields suitable for the present invention. For those familiar with the field, additional modifications can be easily achieved.

Claims

1. A method for accurately identifying the quality of Momordica grosvenori varieties, characterized in that: include: Step 1, testing the quality indicators of mogrosides of several varieties to obtain mogroside content data and sweetness data, wherein the mogroside content data at least includes four main key content indicators: 11-O-mogroside V, mogroside V, isomogroside V, and simanoside I, as well as a total saponin content indicator; Step 2, clustering the acquired data on the content and sweetness of mogrosides to generate a number of cluster categories; Step 3, establishing the corresponding relationship between the quality of Momordica grosvenori varieties and the content of Momordica grosvenori saponins and the clustering categories of sweetness; Step 4: Based on the established corresponding relationship, the variety quality typing of the tested monk fruit samples is performed.

2. The method for accurately identifying the quality of Momordica grosvenori varieties according to claim 1, characterized in that: The step 1 comprises: Sample collection: Collect monk fruit samples from multiple varieties and clearly mark the variety information; Saponin extraction and detection: Extraction and quantitative analysis of mogroside saponins in each sample; Data recording: record the content data of each saponin detected, and calculate the total saponin content index and sweetness index to form a complete data set of monk fruit saponin content and sweetness; Among them, total saponin content = the sum of the contents of 11-O-mogroside V, mogroside V, isomogroside V, and simanoside I; sweetness = (simanoside I content × 465 + isomogroside V content × 500 + mogroside V content × 378 + 11-O-mogroside V content × 68) / 5.

3. The method for accurately identifying the quality of Momordica grosvenori varieties according to claim 2, characterized in that: The mogroside content data also includes any one or more of the indicators of mogroside IIE, mogroside III, mogroside IV, and mogroside IVa.

4. The method for accurately identifying the quality of Momordica grosvenori varieties according to claim 1, characterized in that: The step 2 comprises: Data preprocessing: The data of mogroside content and sweetness were processed using standardized methods; Select clustering algorithm: Select a suitable clustering algorithm based on the characteristics of the data and clustering requirements; Set clustering parameters: set the number of clusters K, metric, and minimum number of samples; Perform clustering processing: input the pre-processed momordica saponin content and sweetness data into the clustering algorithm, perform clustering processing, and generate a number of cluster categories.

5. The method for accurately identifying the quality of Momordica grosvenori varieties according to claim 4, characterized in that: The clustering algorithm selected in the step of selecting a clustering algorithm is a K-means clustering algorithm.

6. The method for accurately identifying the quality of Momordica grosvenori varieties according to claim 4, characterized in that: In step 3, according to the clustering results, a correspondence is established between the quality of the monk fruit varieties and the corresponding clustering categories to form a correspondence table between the quality of the varieties and the categories.

7. The method for accurately identifying the quality of Momordica grosvenori varieties according to claim 6, characterized in that: In step 4, in the step of accurate identification of variety quality typing, the saponin content and sweetness data of the tested monk fruit sample are compared with the established corresponding relationship to determine the variety quality type of the tested monk fruit sample.

8. A model for accurate identification of monk fruit variety quality typing, the model is constructed based on the method for accurate identification of monk fruit variety quality typing according to any one of claims 1 to 7, and is used for accurate identification of monk fruit variety quality typing on monk fruit samples.

9. The accurate identification model for the quality typing of Momordica grosvenori varieties as claimed in claim 8, characterized in that: The model sends all possible variety quality typing and identification results to the user, and the user determines the variety quality based on the results.

10. The accurate identification model for the quality typing of Momordica grosvenori varieties as claimed in claim 8, characterized in that: The model can improve the accuracy of typing and identification as the number of saponin types detected increases.

Images