Honey variety identification method based on amino acid and polyphenol compounds and application of honey variety identification method
By measuring the content of amino acids and polyphenol compounds in honey, combining partial least squares discrimination analysis and random forest algorithm, a honey variety identification model was constructed, which solved the problem of identifying multiple honey varieties in the existing technology, and achieved efficient and accurate identification of honey varieties.
Patent Information
- Application Number
- CN202510911580.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-02
- Publication Date
- 2025-09-02
AI Technical Summary
The existing honey variety identification methods are difficult to achieve accurate and rapid identification when there are many types of samples, and are highly limited and cannot effectively distinguish a variety of single nectars.
The identification method based on amino acids and polyphenol compounds was used to determine the content of amino acids and polyphenol compounds in 14 single nectars through a fully automatic amino acid analyzer and ultra-high performance liquid chromatography-mass spectrometer. A honey variety identification model was constructed based on partial least squares discriminant analysis and random forest algorithm, and a characteristic differential substance was selected, and a Fisher linear discriminant analysis and random forest prediction model was established.
The high accuracy of 14 honey varieties was achieved. The Fisher linear discrimination model's discrimination accuracy was 100%, the cross-validation rate was 98.6%, and the prediction accuracy of the random forest prediction model training set and test set were 100%.
Smart Images
Figure CN120577461A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of honey variety detection, in particular to a honey variety identification method based on amino acids and polyphenol compounds and application thereof. Background Art
[0002] Honey is a beloved sweet food made by bees through a brewing process. Honey is made from nectar and other secretions collected from various plants and then combined with substances produced by their own metabolism. Its main components are sugars, but it is also rich in various bioactive ingredients, such as amino acids, polyphenols, and minerals.
[0003] Honey can be divided into monofloral and polyfloral honeys based on their source. The difference between these two types is whether bees source honey from a single plant or from a variety of plants. Nature offers a vast array of botanical resources for bees to harvest, notably rapeseed, acacia, and linden. Monofloral honeys from various varieties offer high purity, concentrated nutrients, and retain the unique flavor and composition of each plant. Therefore, the purity and high quality of monofloral honey sources have become a key concern for consumers. However, with increasing consumer demand, the favorable supply and demand environment in the honey market has been disrupted.
[0004] Currently, research on the identification and authentication of honey varieties focuses on qualitative and quantitative analysis of one or more chemical components in different honeys to identify differences between honey varieties, or on the use of single chemical components in honey combined with multivariate statistical analysis to discriminate between varieties. However, these existing methods have certain specificity and limitations. They only measure the content of individual target compounds to identify differences between honey varieties, or only measure a single target compound in combination with multivariate statistical analysis to identify varietal differences. This approach can achieve identification goals when the sample type is small, but its limitations become apparent once the sample type increases.
[0005] Therefore, researching and developing a method for identifying different honey varieties and achieving accurate and rapid identification of multiple honey varieties has become a key issue that the industry urgently needs to break through.
[0006] In view of this, the present invention is proposed. Summary of the Invention
[0007] The first purpose of the present invention is to provide a honey variety identification method based on amino acids and polyphenol compounds, so as to achieve accurate and rapid identification of multiple varieties of honey and solve the current limitation problem of honey variety identification based on a single type of indicator.
[0008] In order to achieve the above-mentioned purpose of the present invention, the following technical solutions are adopted: The present invention provides a method for identifying honey varieties based on amino acids and polyphenol compounds, and the identification method comprises the following steps: (I) providing a honey sample to be tested, and determining the content of characteristic amino acids and characteristic polyphenolic compounds in the sample to be tested; The characteristic amino acids include proline (Pro), tyrosine (Tyr), cysteine (Cys), methionine (Met) and aspartic acid (Asp); the characteristic polyphenolic compounds include trans-cinnamic acid, protocatechuic acid, 3,4-dimethylcinnamic acid, rosmarinic acid, apigenin, vanillic acid, coumaric acid, isoferulic acid, caffeic acid phenethyl ester, gallic acid, naringenin, p-coumaric acid, ferulic acid, styraxic acid, chrysin and syringic acid; (II) Constructing a honey variety identification model, introducing the contents of characteristic amino acids and characteristic polyphenolic compounds measured in the honey sample to be tested into the honey variety identification model, and identifying the honey variety indicating the honey sample to be tested.
[0009] Furthermore, there are 14 honey varieties in step (II), including: wolfberry honey, locust flower honey, jujube flower honey, rapeseed honey, linden honey, astragalus honey, wolfberry honey, osmanthus honey, fennel honey, chrysanthemum honey, litchi honey, vitex honey, astragalus honey or loquat honey.
[0010] Furthermore, the method for constructing the honey variety identification model includes: (A) Feature screening: S1: Collect and detect the data of amino acids and polyphenol compounds of 14 honey varieties as the first training dataset; The first training data set includes detection data of 17 amino acids and 24 polyphenolic compounds, and the 17 amino acids and 24 polyphenolic compounds are recorded as first quantitative species difference components; The 17 amino acids are aspartic acid (Asp), threonine (Thr), serine (Ser), glutamic acid (Glu), glycine (Gly), alanine (Ala), cysteine (Cys), valine (Val), methionine (Met), isoleucine (Ile), leucine (Leu), tyrosine (Tyr), phenylalanine (Phe), histidine (His), lysine (Lys), arginine (Arg), and proline (Pro). The 24 polyphenolic compounds are chrysin, gallic acid, galangin, caffeic acid, p-coumaric acid, apigenin, rosmarinic acid, rutin, myricic acid, quercetin, coumaric acid, syringic acid, naringenin, morin, protocatechuic acid, squidrin, ferulic acid, vanillic acid, trans-cinnamic acid, 3,4-dimethylcinnamic acid, caffeic acid phenethyl ester, kaempferol, schizonepeta tenuifolia, and isoferulic acid; S2: Use partial least squares discriminant analysis to reduce the dimensionality of the first training data set to obtain the second training data set; The second training data set includes detection data of 5 amino acids and 16 polyphenolic compounds, and the 5 amino acids and 16 polyphenolic compounds are recorded as the second quantitative species difference components; Among them, the five amino acids are proline (Pro), tyrosine (Tyr), cysteine (Cys), methionine (Met) and aspartic acid (Asp); the 16 polyphenol compounds are trans-cinnamic acid, protocatechuic acid, 3,4-dimethylcinnamic acid, rosmarinic acid, apigenin, vanillic acid, coumaric acid, isoferulic acid, caffeic acid phenethyl ester, gallic acid, naringenin, p-coumaric acid, ferulic acid, styrosinic acid, chrysin and syringic acid; (B) Model construction: Using the random forest algorithm, multiple decision trees are trained using the second training dataset to construct a random forest model as a honey variety identification model.
[0011] Furthermore, the dimensionality reduction processing method of step S2 includes: using partial least squares discriminant analysis to analyze the first training data set, screening amino acid and polyphenol compound data with VIP values greater than 1 generated by the partial least squares discriminant analysis as characteristic difference component data, and obtaining a second training data set.
[0012] Furthermore, in the model construction of step (B), each decision tree is trained by randomly selecting a sample data each time and selecting the characteristic difference component data of the second training data set with replacement. The training is repeated multiple times by bootstrap resampling to obtain 100 decision trees, and the 100 decision trees are used to form a random forest model as a honey variety identification model; the honey variety identification model is used to evaluate the decision results of all decision trees, and the final identification result is output based on the majority principle.
[0013] Furthermore, in the model construction of step (B), for each decision tree, the second quantity type difference component data is selected with replacement as the sample at the root node of the decision tree, all available second quantity type difference components are used for evaluation when each node of the decision tree needs to be split, and the second quantity type difference component that minimizes the node Gini coefficient is selected as the splitting condition of the node of the decision tree; based on the selected second quantity type difference component, a splitting operation is performed on each node so that each child node contains a part of the sample data; and this process is repeated until each leaf node contains at least 3 samples.
[0014] Furthermore, the sample data of multiple honey varieties obtained by data collection were divided into data according to a certain ratio, wherein 80% of the data were used as a training set and 20% of the data were used as a test set; stratified sampling was used to ensure that the proportion of each category in the training set and the test set was reasonable; the synthetic minority group oversampling technology was applied to the training set to deal with the category imbalance problem and maintain the original data distribution of the test set; the training set data was used for feature screening, model construction and optimization, and the constructed honey variety identification model was tested using the test set to evaluate the accuracy of the honey variety identification model on an independent data set.
[0015] Furthermore, the model construction in step (B) can be replaced by: using the linear discriminant analysis method, using the second training data set as the training set to establish a discriminant function and construct a honey variety identification model.
[0016] The present invention provides an application of the honey variety identification method based on amino acids and polyphenol compounds in honey variety identification.
[0017] It should be noted that there are many types of single-flower honey, as many as 14. Existing honey variety identification methods are difficult to distinguish the types of single-flower honey from such a wide variety.
[0018] In view of this, the present invention uses a fully automatic amino acid analyzer and an ultra-high performance liquid chromatography-mass spectrometry to determine the contents of amino acids and polyphenol compounds in 14 types of single-flower honey, including wolfberry honey, locust flower honey, jujube flower honey, rapeseed honey, linden honey, astragalus honey, wolfberry honey, osmanthus honey, fennel honey, chrysanthemum honey, litchi honey, vitex honey, milk vetch honey and loquat honey. Subsequently, a multivariate statistical analysis method is comprehensively used to establish a classification prediction model based on the random forest algorithm, explore the feasibility of honey variety discrimination, and provide a reference basis for honey variety traceability.
[0019] Specifically, in order to achieve high accuracy in model discrimination, the present invention uses partial least squares discriminant analysis to statistically analyze the contents of amino acids and polyphenolic compounds in 14 types of single flower nectars and screen out characteristic differential substances (trans-cinnamic acid, protocatechuic acid, Pro, 3,4-dimethylcinnamic acid, rosmarinic acid, apigenin, Tyr, vanillic acid, coumaric acid, isoferulic acid, Cys, caffeic acid phenethyl ester, Met, Asp, gallic acid, naringenin, p-coumaric acid, ferulic acid, schizonepeta tenuifolia, chrysin and syringic acid) for the establishment of Fisher linear discriminant analysis and random forest prediction models.
[0020] The constructed Fisher linear discriminant model achieved 100% accuracy and a cross-validation rate of 98.6%. The random forest prediction model also achieved 100% accuracy on both the training and test sets. This demonstrates that the random forest prediction model performs better than the Fisher linear discriminant model.
[0021] Compared with the prior art, the present invention has the following beneficial effects: The invention provides a honey variety identification method based on amino acids and polyphenolic compounds. The identification method adopts a PLS-DA analysis method to analyze characteristic amino acids and characteristic polyphenolic compounds of 14 honey varieties, screens out characteristic difference index components with VIP values greater than 1 (trans-cinnamic acid, protocatechuic acid, Pro, 3,4-dimethylcinnamic acid, rosmarinic acid, apigenin, Tyr, vanillic acid, coumaric acid, isoferulic acid, Cys, caffeic acid phenethyl ester, Met, Asp, gallic acid, naringenin, p-coumaric acid, ferulic acid, styrosinic acid, chrysin and syringic acid), and then uses the data of the characteristic amino acids and characteristic polyphenolic compounds after the screening for Fisher linear discriminant analysis and the establishment of a random forest prediction model.
[0022] It has been verified that the constructed Fisher linear discriminant model has a discrimination accuracy of 100% and a cross-validation rate of 98.6%. The prediction accuracy of the random forest prediction model training set and test set are both 100%, which can be used accurately and efficiently to identify honey varieties. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0024] Figure 1 The standard spectrum of 17 amino acids in honey provided in Example 1 of the present invention; Figure 2 The standard spectrum of 24 polyphenol compounds in honey provided in Example 1 of the present invention; Figure 3 This is a score graph of the PLS-DA analysis provided in Example 2 of the present invention; Figure 4 VIP diagram of the PLS-DA analysis provided in Example 2 of the present invention; Figure 5 This is a permutation validation diagram of the PLS-DA analysis provided in Example 2 of the present invention; Figure 6 A scatter plot showing the classification of different honey varieties provided in Example 3 of the present invention; Figure 7 This is a training set prediction accuracy graph provided by Example 4 of the present invention; Figure 8This is a test set prediction accuracy graph provided by Example 4 of the present invention; Figure 9 The confusion matrix diagram of the training set provided in Example 4 of the present invention; Figure 10 This is the confusion matrix diagram of the test set provided by Example 4 of the present invention. DETAILED DESCRIPTION
[0025] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.
[0027] The technical solutions of the present invention will be further described below with reference to the examples. The experimental methods in the following examples are conventional methods unless otherwise specified. The materials, reagents, etc. used in the following examples are all commercially available unless otherwise specified.
[0028] Example 1 Determination of amino acids and polyphenols in different honey varieties (1) Sample collection: The experiment collected 14 types of single-flower honey (wolfberry honey, locust flower honey, jujube flower honey, rapeseed honey, linden honey, astragalus honey, wolfberry honey, osmanthus honey, fennel honey, chrysanthemum honey, litchi honey, vitex honey, milk vetch honey, and loquat honey) from large supermarkets, farmers' markets, and production enterprises from 2023 to 2024 (all products must be marked with the origin mark or obtained directly from the production enterprises in the place of origin) and stored at 4°C.
[0029] (2) Determination of amino acids: 1) Sample Pretreatment: Accurately weigh 200 mg of honey sample (accurate to 0.001 g) and place it in a 25 mL hydrolysis tube. Add 10 mL of 6 mol / L hydrochloric acid solution. After thorough mixing, vortex the sample for 3 minutes. The sample was then placed in a 110°C electric constant-temperature drying oven for 24 hours of hydrolysis. After hydrolysis, the sample was removed and concentrated to near dryness using a nitrogen purge apparatus. The sample was then diluted to 1 mL with the pre-prepared sample diluent. After centrifugation at 20,000 rpm for 3 minutes, the sample was filtered through a 0.22 µm microporous membrane and analyzed.
[0030] 2) Preparation of standard solutions: Use a pipette to accurately measure 1 mL of the mixed standard solution and transfer it to a 100 mL volumetric flask. Add the sample diluent to the mark to prepare a mixed standard intermediate solution with a concentration of 1 µmol / mL. Subsequently, accurately pipette an appropriate amount of this intermediate solution (1 µmol / mL) into a 10 mL volumetric flask and dilute to volume with the sample diluent to prepare a series of standard working solutions with a concentration gradient of 25, 50, 60, 80, 100, 200, and 250 nmol / mL.
[0031] 3) Instrument conditions: The fully automatic amino acid analyzer was equipped with a diode photometer with detection wavelengths of 570 and 440 nm. The chromatographic column was a cation exchange column from Na column Kit HR. The column temperature was 50°C (0–35 min), 50–60°C (35–37 min), 60°C (37–52 min), 74°C (52–55 min), 74°C (55–74 min), 48°C (74–79 min), and 48°C (79–90 min). The injection volume was 20 µL. The mobile phase was 100% A (0–25 min), 100% B (25–36 min), 100% C (36–53 min), 100% D (53–72 min), 100% F (72–79 min), and 100% A (79–90 min). The flow rate was 0.4 mL / min. The post-column reaction cell temperature was 115°C.
[0032] (3) Amino acid determination results of 14 different types of honey The detection wavelength for aspartic acid (Asp), threonine (Thr), serine (Ser), glutamic acid (Glu), glycine (Gly), alanine (Ala), cysteine (Cys), valine (Val), methionine (Met), isoleucine (Ile), leucine (Leu), tyrosine (Tyr), phenylalanine (Phe), histidine (His), lysine (Lys), and arginine (Arg) is 570 nm, and the detection wavelength for proline (Pro) is 440 nm. The standard spectrum of amino acids is as follows: Figure 1 shown.
[0033] Figure 1 This example provides a standard profile of 17 amino acids in honey.
[0034] The total amino acid content of 14 different honey varieties ranged from 0.0996% to 0.8600%. The top five honeys were fennel honey, lychee honey, vitex honey, astragalus honey, and wolftooth honey. Fennel honey had the highest total amino acid content, at 0.8600%. Rapeseed honey had a much lower amino acid content than the other honeys, at 0.0996%. Arginine was not detected in any of the honeys, indicating that it is likely not present in natural honeys, as shown in Tables 1 and 2.
[0035] Table 1:
[0036] Table 2:
[0037] As shown in Tables 1 and 2, the contents of aspartic acid, glutamic acid, valine and leucine in the 14 honey samples are generally at high levels, and this distribution feature is similar to the amino acid composition in pollen.
[0038] (IV) Determination of polyphenolic compounds 1) Sample Pretreatment: Accurately weigh 5 g of honey sample (accurate to 0.001 g) and place it in a 50 mL centrifuge tube. Add 4 mL of ultrapure water and thoroughly stir using a vortex mixer for 5 minutes to ensure complete dissolution of the honey. Ultrasonic extraction was then performed for 10 minutes. Finally, the solution was brought to a volume of 10 mL with ultrapure water. The solution was filtered through a 0.22 μm microporous membrane before analysis.
[0039] 2) Preparation of Standard Solutions: Accurately weigh appropriate amounts of standard samples and dissolve them in methanol to prepare 1.0 mg / mL stock solutions of individual standards (store at -20°C). Then, accurately transfer 100 µL of each stock solution to a 100 mL volumetric flask and dilute to the mark with methanol to prepare a 1 µg / mL mixed standard intermediate solution. Using the initial mobile phase, prepare mixed standard working solutions at concentrations of 0.1, 0.2, 0.5, 1.0, 2.0, 5.0, 10.0, 20.0, and 50.0 ng / mL, respectively.
[0040] 3) Instrument conditions: Chromatographic conditions: Column: Thermo Scientific C18 column (1.9 µm, 2.1 mm × 100 mm); Column temperature: 30°C; Injection volume: 5 µL; Mobile phase: A: 5 mmol / L ammonium acetate in 0.1% formic acid; B: acetonitrile; Flow rate: 0.25 mL / min; Gradient elution program: 0–1.0 min, 95% A; 1.0–10.0 min, 95%–10% A; 10.0–12.0 min, 10% A; 12.0–12.1 min, 10%–95% A; 12.1–15.0 min, 95% A. Mass spectrometry conditions: Electrospray ionization (ESI) source, negative ion mode; ionization voltage (IS): -4000 V; spray gas pressure (GS1): 50 psi; auxiliary heater gas pressure (GS2): 55 psi; curtain gas pressure (CUR): 35 psi; auxiliary heater gas temperature (TEM): 500°C; dynamic multiple reaction monitoring (MRM) scanning mode. The mass spectrometry parameters for the 24 polyphenols are shown in Table 3.
[0041] Table 3:
[0042] (V) Determination of polyphenolic compounds in 14 different types of honey Figure 2 This example provides a standard spectrum of 24 polyphenol compounds in honey.
[0043] The contents of 24 polyphenolic compounds in 14 different varieties of honey are shown in Tables 4 and 5.
[0044] Table 4:
[0045] Table 5:
[0046] As shown in Tables 4 and 5, significant differences in polyphenolic compounds were observed among different honey varieties, with total polyphenolic content ranging from 15.450 μg / kg to 117.388 μg / kg. Jujube honey contained significantly higher levels of polyphenolic compounds than other honey varieties, followed by wolfberry honey and astragalus honey. Loquat honey had the lowest polyphenol content. Jujube honey also contained higher levels of p-coumaric acid, syringic acid, ferulic acid, vanillic acid, and kaempferol than other honey varieties. Astragalus honey had the highest levels of caffeic acid and coumaric acid, reaching 11.842 μg / kg and 8.078 μg / kg, respectively. Vitex honey also contained the highest levels of protocatechuic acid, reaching 18.359 μg / kg.
[0047] Example 2 Partial least squares discriminant analysis (PLS-DA) is a supervised multivariate statistical technique that separates variable data and categorical information into two independent datasets. It then uses projection and discriminant calculations to analyze the data, achieving ideal separation between groups and identifying variables that contribute significantly to classification. In this example, PLS-DA was used to establish a correlation model between the amino acid and polyphenol content of different honey varieties, achieving better classification results and identifying important variables that influence classification.
[0048] Specifically, this example uses the contents of 17 amino acids and 24 polyphenolic compounds in 14 different varieties of honey obtained in Example 1 to perform PLS-DA analysis. The results are shown in Figures 3 to 5 .
[0049] Figure 3 is the score graph of PLS-DA analysis in this embodiment; wherein, Figure 3 A in the middle is wolfberry honey, Figure 3 B is Sophora japonica honey, Figure 3 C is jujube honey, Figure 3 D in the middle is rapeseed honey, Figure 3 E in the middle is linden honey, Figure 3 F is Astragalus honey, Figure 3 G in the middle is wolfberry honey, Figure 3 H in the middle is osmanthus honey, Figure 3 I is fennel honey, Figure 3 J is chrysanthemum honey, Figure 3 K in the middle is lychee honey, Figure 3 L in the middle is vitex honey, Figure 3 M in the middle is milk vetch honey, Figure 3 The N in it stands for loquat honey.
[0050] Figure 4 VIP diagram of the PLS-DA analysis in this example; Figure 5 This is the permutation validation diagram of the PLS-DA analysis in this example; like Figure 3The PLS-DA score graph shows that different varieties of honey have a good separation effect. Different varieties are relatively clustered and can be clearly distinguished. Only some samples of milk vetch honey are closer to loquat honey. Astragalus honey and chrysanthemum honey are relatively compactly clustered in the same area, but can be distinguished from each other.
[0051] See also Figure 4 VIP values generated by PLS-DA were used to identify distinctive compounds (VIP > 1). Higher VIP values indicate a more significant role for the compound in distinguishing between sample groups. In this study, compounds with VIP values greater than 1 included trans-cinnamic acid, protocatechuic acid, Pro, 3,4-dimethylcinnamic acid, rosmarinic acid, apigenin, Tyr, vanillic acid, coumaric acid, isoferulic acid, Cys, caffeic acid phenethyl ester, Met, Asp, gallic acid, naringenin, p-coumaric acid, ferulic acid, styrosinic acid, chrysin, and syringic acid.
[0052] The established PLS-DA model was validated 200 times, and the results are shown in Figure 5 . Figure 5 The R² and Q² values on the far right of the graph both exceed 0.5, and the original data on the right are significantly higher than the Q² and R² values on the left. Furthermore, the intersection of the Q² regression line and the vertical axis lies in the negative region. This indicates that the constructed PLS-DA model is not overfitting and exhibits good predictive performance.
[0053] Example 3 Linear discriminant analysis (LDA) is a statistical technique used to determine the classification of samples. This method constructs a linear discriminant function using the observed values of multiple variables in known samples to classify unknown samples. Furthermore, LDA's cross-validation strategy uses all samples other than a specific sample as a training set to construct a discriminant function and classify the sample, thereby assessing the reliability of the discriminant model.
[0054] Specifically, this embodiment uses the Fisher discriminant function to construct a universal identification model for multivariate statistical analysis of various honey samples. The characteristic difference components with VIP values greater than 1 in PLS-DA (trans-cinnamic acid, protocatechuic acid, Pro, 3,4-dimethylcinnamic acid, rosmarinic acid, apigenin, Tyr, vanillic acid, coumaric acid, isoferulic acid, Cys, caffeic acid phenethyl ester, Met, Asp, gallic acid, naringenin, p-coumaric acid, ferulic acid, styrosinic acid, chrysin, and syringic acid) are selected as independent variables for stepwise discriminant analysis.
[0055] The extraction model obtained 13 typical discriminant functions, among which Y1~Y 13represents the first through 13th typical discriminant functions, respectively, and X represents the 21 identified differential features. The first and second discriminant functions contributed 46.8% and 27.5% of the variance, respectively, for a cumulative contribution of 74.3%. A scatter plot of the classification of different honey varieties was plotted, with the first and second discriminant functions as the horizontal and vertical coordinates.
[0056] The 13 typical discriminant functions are as follows: Y1=0.787X Asp -128.195X Cys -5.253X Met +21.435X Tyr +32.853X Pro -18.886X 白杨素 +0.552X 没食子酸 +1.468X 对香豆酸 -0.839X 芹菜素 +0.137X 迷迭香酸 -0.644X 香豆酸 +0.675X 丁香酸 -32.317X 柚皮素 +0.141X 原儿茶酸 +0.746X 阿魏酸 -0.454X 香草酸 +0.221X 反式肉桂酸 +0.356X 3,4-二甲基肉桂酸 +2.099X 咖啡酸苯乙基酯 +4.828X 犀草酸 +1.185X 异阿魏酸 -11.5.
[0057] Y2=23.641X Asp +457.865X Cys +322.764X Met +154.531X Tyr +188.5X Pro +28.107X 白杨素 +0.466X 没食子酸 +0.196X 对香豆酸 -1.017X 芹菜素 -0.319X 迷迭香酸 +2.046X 香豆酸 -0.284X 丁香酸 +40.749X 柚皮素 -0.639X 原儿茶酸 +0.302X 阿魏酸 +0.253X 香草酸 +1.49X 反式肉桂酸 -1.263X 3,4-二甲基肉桂酸<h2 style=";text-align:left;direction:ltr">-36.228X<h2 style=";text-align:left;direction:ltr"> 咖啡酸苯乙基酯 <h2 style=";text-align:left;direction:ltr"> +1.951X<h2 style=";text-align:left;direction:ltr"> 犀草酸 <h2 style=";text-align:left;direction:ltr"> +2.162X<h2 style=";text-align:left;direction:ltr"> 异阿魏酸 <h2 style=";text-align:left;direction:ltr"> -20.631.<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">
[0058] <h2 style=";text-align:left;direction:ltr"> Y3=7.975X<h2 style=";text-align:left;direction:ltr"> Asp <h2 style=";text-align:left;direction:ltr"> +406.713X<h2 style=";text-align:left;direction:ltr"> Cys <h2 style=";text-align:left;direction:ltr"> -7.963X<h2 style=";text-align:left;direction:ltr"> Met <h2 style=";text-align:left;direction:ltr"> -36.73X<h2 style=";text-align:left;direction:ltr"> Tyr <h2 style=";text-align:left;direction:ltr"> +132.176X<h2 style=";text-align:left;direction:ltr"> Pro <h2 style=";text-align:left;direction:ltr"> +1.028X<h2 style=";text-align:left;direction:ltr"> 白杨素 <h2 style=";text-align:left;direction:ltr"> +0.304X<h2 style=";text-align:left;direction:ltr"> 没食子酸 <h2 style=";text-align:left;direction:ltr"> -1.084X<h2 style=";text-align:left;direction:ltr"> 对香豆酸 <h2 style=";text-align:left;direction:ltr"> -0.771X<h2 style=";text-align:left;direction:ltr"> 芹菜素 <h2 style=";text-align:left;direction:ltr"> -1.473X<h2 style=";text-align:left;direction:ltr"> 迷迭香酸 <h2 style=";text-align:left;direction:ltr"> +0.185X<h2 style=";text-align:left;direction:ltr"> 香豆酸 <h2 style=";text-align:left;direction:ltr"> +0.475X<h2 style=";text-align:left;direction:ltr"> 丁香酸 <h2 style=";text-align:left;direction:ltr"> -33.331X<h2 style=";text-align:left;direction:ltr"> 柚皮素 <h2 style=";text-align:left;direction:ltr"> +0.343X<h2 style=";text-align:left;direction:ltr"> 原儿茶酸 <h2 style=";text-align:left;direction:ltr"> +0.412X<h2 style=";text-align:left;direction:ltr"> 阿魏酸 <h2 style=";text-align:left;direction:ltr"> +0.49X<h2 style=";text-align:left;direction:ltr"> 香草酸 <h2 style=";text-align:left;direction:ltr"> +0.865X<h2 style=";text-align:left;direction:ltr"> 反式肉桂酸 <h2 style=";text-align:left;direction:ltr"> -1.952X<h2 style=";text-align:left;direction:ltr"> 3,4-二甲基肉桂酸 <h2 style=";text-align:left;direction:ltr"> -3.687X<h2 style=";text-align:left;direction:ltr"> 咖啡酸苯乙基酯 <h2 style=";text-align:left;direction:ltr"> -21.464X<h2 style=";text-align:left;direction:ltr"> 犀草酸 <h2 style=";text-align:left;direction:ltr"> +0.798X<h2 style=";text-align:left;direction:ltr"> 异阿魏酸 <h2 style=";text-align:left;direction:ltr"> -4.951.<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">
[0059] <h2 style=";text-align:left;direction:ltr"> Y4=-2.853X<h2 style=";text-align:left;direction:ltr"> Asp <h2 style=";text-align:left;direction:ltr"> -224.76X<h2 style=";text-align:left;direction:ltr"> Cys <h2 style=";text-align:left;direction:ltr"> -29.171X<h2 style=";text-align:left;direction:ltr"> Met <h2 style=";text-align:left;direction:ltr"> -125.72X<h2 style=";text-align:left;direction:ltr"> Tyr <h2 style=";text-align:left;direction:ltr"> -80.271X<h2 style=";text-align:left;direction:ltr"> Pro <h2 style=";text-align:left;direction:ltr"> -25.447X<h2 style=";text-align:left;direction:ltr"> 白杨素 <h2 style=";text-align:left;direction:ltr"> +2.82X<h2 style=";text-align:left;direction:ltr"> 没食子酸 <h2 style=";text-align:left;direction:ltr"> -0.518X<h2 style=";text-align:left;direction:ltr"> 对香豆酸 <h2 style=";text-align:left;direction:ltr"> -3.081X<h2 style=";text-align:left;direction:ltr"> 芹菜素 <h2 style=";text-align:left;direction:ltr"> +1.689X<h2 style=";text-align:left;direction:ltr"> 迷迭香酸 <h2 style=";text-align:left;direction:ltr"> +0.02X<h2 style=";text-align:left;direction:ltr"> 香豆酸 <h2 style=";text-align:left;direction:ltr"> +0.503X<h2 style=";text-align:left;direction:ltr"> 丁香酸 <h2 style=";text-align:left;direction:ltr"> +20.15X<h2 style=";text-align:left;direction:ltr"> 柚皮素 <h2 style=";text-align:left;direction:ltr"> +0.355X<h2 style=";text-align:left;direction:ltr"> 原儿茶酸 <h2 style=";text-align:left;direction:ltr"> +0.798X<h2 style=";text-align:left;direction:ltr"> 阿魏酸 <h2 style=";text-align:left;direction:ltr"> +0.505X<h2 style=";text-align:left;direction:ltr"> 香草酸 <h2 style=";text-align:left;direction:ltr"> -1.3X<h2 style=";text-align:left;direction:ltr"> 反式肉桂酸 <h2 style=";text-align:left;direction:ltr"> -1.139X<h2 style=";text-align:left;direction:ltr"> 3,4-二甲基肉桂酸 <h2 style=";text-align:left;direction:ltr"> -23.833X<h2 style=";text-align:left;direction:ltr"> 咖啡酸苯乙基酯 <h2 style=";text-align:left;direction:ltr"> +20.889X<h2 style=";text-align:left;direction:ltr"> 犀草酸 <h2 style=";text-align:left;direction:ltr"> -1.118X<h2 style=";text-align:left;direction:ltr"> 异阿魏酸 <h2 style=";text-align:left;direction:ltr"> -3.522.<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">
[0060] <h2 style=";text-align:left;direction:ltr"> Y5=23.007X<h2 style=";text-align:left;direction:ltr"> Asp <h2 style=";text-align:left;direction:ltr"> -313.162X<h2 style=";text-align:left;direction:ltr"> Cys <h2 style=";text-align:left;direction:ltr"> -796.384X<h2 style=";text-align:left;direction:ltr"> Met <h2 style=";text-align:left;direction:ltr"> -48.472X<h2 style=";text-align:left;direction:ltr"> Tyr <h2 style=";text-align:left;direction:ltr"> +125.216X<h2 style=";text-align:left;direction:ltr"> Pro <h2 style=";text-align:left;direction:ltr"> +26.629X<h2 style=";text-align:left;direction:ltr"> 白杨素 <h2 style=";text-align:left;direction:ltr"> -2.597X<h2 style=";text-align:left;direction:ltr"> 没食子酸 <h2 style=";text-align:left;direction:ltr"> -0.063X<h2 style=";text-align:left;direction:ltr"> 对香豆酸 <h2 style=";text-align:left;direction:ltr"> +0.901X<h2 style=";text-align:left;direction:ltr"> 芹菜素 <h2 style=";text-align:left;direction:ltr"> -0.842X<h2 style=";text-align:left;direction:ltr"> 迷迭香酸 <h2 style=";text-align:left;direction:ltr"> -1.692X<h2 style=";text-align:left;direction:ltr"> 香豆酸 <h2 style=";text-align:left;direction:ltr"> +0.205X<h2 style=";text-align:left;direction:ltr"> 丁香酸 <h2 style=";text-align:left;direction:ltr"> +1.89X<h2 style=";text-align:left;direction:ltr"> 柚皮素 <h2 style=";text-align:left;direction:ltr"> -0.233X<h2 style=";text-align:left;direction:ltr"> 原儿茶酸 <h2 style=";text-align:left;direction:ltr"> -0.53X<h2 style=";text-align:left;direction:ltr"> 阿魏酸 <h2 style=";text-align:left;direction:ltr"> -0.194X<h2 style=";text-align:left;direction:ltr"> 香草酸 <h2 style=";text-align:left;direction:ltr"> +3.467X<h2 style=";text-align:left;direction:ltr"> 反式肉桂酸 <h2 style=";text-align:left;direction:ltr"> +0.763X<h2 style=";text-align:left;direction:ltr"> 3,4-二甲基肉桂酸 <h2 style=";text-align:left;direction:ltr"> +56.397X<h2 style=";text-align:left;direction:ltr"> 咖啡酸苯乙基酯 <h2 style=";text-align:left;direction:ltr"> +19.203X<h2 style=";text-align:left;direction:ltr"> 犀草酸 <h2 style=";text-align:left;direction:ltr"> -0.232X<h2 style=";text-align:left;direction:ltr"> 异阿魏酸 <h2 style=";text-align:left;direction:ltr"> +3.118.<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">
[0061] <h2 style=";text-align:left;direction:ltr"> Y6=13.261X<h2 style=";text-align:left;direction:ltr"> Asp <h2 style=";text-align:left;direction:ltr"> +141.213X<h2 style=";text-align:left;direction:ltr"> Cys <h2 style=";text-align:left;direction:ltr"> +259.778X<h2 style=";text-align:left;direction:ltr"> Met <h2 style=";text-align:left;direction:ltr"> +164.086X<h2 style=";text-align:left;direction:ltr"> Tyr <h2 style=";text-align:left;direction:ltr"> +18.385X<h2 style=";text-align:left;direction:ltr"> Pro <h2 style=";text-align:left;direction:ltr"> +37.52X<h2 style=";text-align:left;direction:ltr"> 白杨素 <h2 style=";text-align:left;direction:ltr"> +0.944X<h2 style=";text-align:left;direction:ltr"> 没食子酸 <h2 style=";text-align:left;direction:ltr"> -0.084X<h2 style=";text-align:left;direction:ltr"> 对香豆酸 <h2 style=";text-align:left;direction:ltr"> -0.435X<h2 style=";text-align:left;direction:ltr"> 芹菜素 <h2 style=";text-align:left;direction:ltr"> -1.245X<h2 style=";text-align:left;direction:ltr"> 迷迭香酸 <h2 style=";text-align:left;direction:ltr"> +0.457X<h2 style=";text-align:left;direction:ltr"> 香豆酸 <h2 style=";text-align:left;direction:ltr"> +0.492X<h2 style=";text-align:left;direction:ltr"> 丁香酸 <h2 style=";text-align:left;direction:ltr"> +7.3X<h2 style=";text-align:left;direction:ltr"> 柚皮素 <h2 style=";text-align:left;direction:ltr"> -0.358X<h2 style=";text-align:left;direction:ltr"> 原儿茶酸 <h2 style=";text-align:left;direction:ltr"> -1.074X<h2 style=";text-align:left;direction:ltr"> 阿魏酸 <h2 style=";text-align:left;direction:ltr"> -0.188X<h2 style=";text-align:left;direction:ltr"> 香草酸 <h2 style=";text-align:left;direction:ltr"> -3.097X<h2 style=";text-align:left;direction:ltr"> 反式肉桂酸 <h2 style=";text-align:left;direction:ltr"> +0.718X<h2 style=";text-align:left;direction:ltr"> 3,4-二甲基肉桂酸 <h2 style=";text-align:left;direction:ltr"> +29.947X<h2 style=";text-align:left;direction:ltr"> 咖啡酸苯乙基酯 <h2 style=";text-align:left;direction:ltr"> +0.322X<h2 style=";text-align:left;direction:ltr"> 犀草酸 <h2 style=";text-align:left;direction:ltr"> +1.594X<h2 style=";text-align:left;direction:ltr"> 异阿魏酸 <h2 style=";text-align:left;direction:ltr"> -7.933.<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">
[0062] <h2 style=";text-align:left;direction:ltr"> Y7=4.381X<h2 style=";text-align:left;direction:ltr"> Asp <h2 style=";text-align:left;direction:ltr"> +378.179X<h2 style=";text-align:left;direction:ltr"> Cys <h2 style=";text-align:left;direction:ltr"> -382.388X<h2 style=";text-align:left;direction:ltr"> Met<h2 style=";text-align:left;direction:ltr">+119.805X<h2 style=";text-align:left;direction:ltr"> Tyr <h2 style=";text-align:left;direction:ltr"> +32.611X<h2 style=";text-align:left;direction:ltr"> Pro <h2 style=";text-align:left;direction:ltr"> -83.358X<h2 style=";text-align:left;direction:ltr"> 白杨素 <h2 style=";text-align:left;direction:ltr"> +0.966X<h2 style=";text-align:left;direction:ltr"> 没食子酸 <h2 style=";text-align:left;direction:ltr"> -0.409X<h2 style=";text-align:left;direction:ltr"> 对香豆酸 <h2 style=";text-align:left;direction:ltr"> +1.829X<h2 style=";text-align:left;direction:ltr"> 芹菜素 <h2 style=";text-align:left;direction:ltr"> +1.962X<h2 style=";text-align:left;direction:ltr"> 迷迭香酸 <h2 style=";text-align:left;direction:ltr"> -0.449X<h2 style=";text-align:left;direction:ltr"> 香豆酸 <h2 style=";text-align:left;direction:ltr"> +0.139X<h2 style=";text-align:left;direction:ltr"> 丁香酸 <h2 style=";text-align:left;direction:ltr"> +4.007X<h2 style=";text-align:left;direction:ltr"> 柚皮素 <h2 style=";text-align:left;direction:ltr"> +0.393X<h2 style=";text-align:left;direction:ltr"> 原儿茶酸 <h2 style=";text-align:left;direction:ltr"> +0.336X<h2 style=";text-align:left;direction:ltr"> 阿魏酸 <h2 style=";text-align:left;direction:ltr"> +0.267X<h2 style=";text-align:left;direction:ltr"> 香草酸 <h2 style=";text-align:left;direction:ltr"> -1.61X<h2 style=";text-align:left;direction:ltr"> 反式肉桂酸 <h2 style=";text-align:left;direction:ltr"> +1.647X<h2 style=";text-align:left;direction:ltr"> 3,4-二甲基肉桂酸 <h2 style=";text-align:left;direction:ltr"> -8.486X<h2 style=";text-align:left;direction:ltr"> 咖啡酸苯乙基酯 <h2 style=";text-align:left;direction:ltr"> +16.479X<h2 style=";text-align:left;direction:ltr"> 犀草酸 <h2 style=";text-align:left;direction:ltr"> +0.274X<h2 style=";text-align:left;direction:ltr"> 异阿魏酸 <h2 style=";text-align:left;direction:ltr"> -9.677.<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">
[0063] <h2 style=";text-align:left;direction:ltr"> Y8=16.1X<h2 style=";text-align:left;direction:ltr"> Asp <h2 style=";text-align:left;direction:ltr"> -151.289X<h2 style=";text-align:left;direction:ltr"> Cys <h2 style=";text-align:left;direction:ltr"> +217.489X<h2 style=";text-align:left;direction:ltr"> Met <h2 style=";text-align:left;direction:ltr"> -73.015X<h2 style=";text-align:left;direction:ltr"> Tyr <h2 style=";text-align:left;direction:ltr"> +4.63X<h2 style=";text-align:left;direction:ltr"> Pro <h2 style=";text-align:left;direction:ltr"> +20.46X<h2 style=";text-align:left;direction:ltr"> 白杨素 <h2 style=";text-align:left;direction:ltr"> +0.507X<h2 style=";text-align:left;direction:ltr"> 没食子酸 <h2 style=";text-align:left;direction:ltr"> +0.151X<h2 style=";text-align:left;direction:ltr"> 对香豆酸 <h2 style=";text-align:left;direction:ltr"> -1.018X<h2 style=";text-align:left;direction:ltr"> 芹菜素 <h2 style=";text-align:left;direction:ltr"> +0.632X<h2 style=";text-align:left;direction:ltr"> 迷迭香酸 <h2 style=";text-align:left;direction:ltr"> -0.519X<h2 style=";text-align:left;direction:ltr"> 香豆酸 <h2 style=";text-align:left;direction:ltr"> -0.12X<h2 style=";text-align:left;direction:ltr"> 丁香酸 <h2 style=";text-align:left;direction:ltr"> -22.146X<h2 style=";text-align:left;direction:ltr"> 柚皮素 <h2 style=";text-align:left;direction:ltr"> +0.52X<h2 style=";text-align:left;direction:ltr"> 原儿茶酸 <h2 style=";text-align:left;direction:ltr"> +1.017X<h2 style=";text-align:left;direction:ltr"> 阿魏酸 <h2 style=";text-align:left;direction:ltr"> -0.622X<h2 style=";text-align:left;direction:ltr"> 香草酸 <h2 style=";text-align:left;direction:ltr"> -0.089X<h2 style=";text-align:left;direction:ltr"> 反式肉桂酸 <h2 style=";text-align:left;direction:ltr"> +0.619X<h2 style=";text-align:left;direction:ltr"> 3,4-二甲基肉桂酸 <h2 style=";text-align:left;direction:ltr"> -25.732X<h2 style=";text-align:left;direction:ltr"> 咖啡酸苯乙基酯 <h2 style=";text-align:left;direction:ltr"> +3.727X<h2 style=";text-align:left;direction:ltr"> 犀草酸 <h2 style=";text-align:left;direction:ltr"> +2.055X<h2 style=";text-align:left;direction:ltr"> 异阿魏酸 <h2 style=";text-align:left;direction:ltr"> -4.345.<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">
[0064] <h2 style=";text-align:left;direction:ltr"> Y9=-12.109X<h2 style=";text-align:left;direction:ltr"> Asp <h2 style=";text-align:left;direction:ltr"> -49.633X<h2 style=";text-align:left;direction:ltr"> Cys <h2 style=";text-align:left;direction:ltr"> -219.814X<h2 style=";text-align:left;direction:ltr"> Met <h2 style=";text-align:left;direction:ltr"> +45.326X<h2 style=";text-align:left;direction:ltr"> Tyr <h2 style=";text-align:left;direction:ltr"> -60.396X<h2 style=";text-align:left;direction:ltr"> Pro <h2 style=";text-align:left;direction:ltr"> +13.291X<h2 style=";text-align:left;direction:ltr"> 白杨素 <h2 style=";text-align:left;direction:ltr"> -0.401X没食子酸 -0.241X 对香豆酸 -0.624X 芹菜素 -1.265X 迷迭香酸 -0.846X 香豆酸 -0.057X 丁香酸 +8.838X 柚皮素 +0.093X 原儿茶酸 +0.842X 阿魏酸 +0.388X 香草酸 +2.934X 反式肉桂酸 +1.901X 3,4-二甲基肉桂酸 +12.364X 咖啡酸苯乙基酯 -5.877X 犀草酸 +0.791X 异阿魏酸 -3.863。
[0065] Y 10 =17.626X Asp -221.298X Cys -587.233X Met +57.398X Tyr -2.15X Pro -2.613X 白杨素 +2.129X 没食子酸 -0.146X 对香豆酸 +2.739X 芹菜素 -0.024X 迷迭香酸 +0.917X 香豆酸 +0.295X 丁香酸 -31.116X 柚皮素 -0.42X 原儿茶酸 +0.484X 阿魏酸 -0.319X 香草酸 +0.004X 反式肉桂酸 +0.454X 3,4-二甲基肉桂酸 +3.883X 咖啡酸苯乙基酯 -3.396X 犀草酸 +0.371X 异阿魏酸 -0.659。
[0066] Y 11 =-4.638X Asp +364.599X Cys +443.143X Met +153.575X Tyr -9.283X Pro -36.463X 白杨素 +1.469X 没食子酸 -0.068X 对香豆酸 -1.754X 芹菜素+2.488X 迷迭香酸 -0.263X 香豆酸 +0.234X 丁香酸 -13.537X 柚皮素 -0.106X 原儿茶酸 -0.816X 阿魏酸 -0.217X 香草酸 +2.645X 反式肉桂酸 -0.029X 3,4-二甲基肉桂酸 -14.411X 咖啡酸苯乙基酯 +19.328X 犀草酸 +0.402X 异阿魏酸 -6.934。
[0067] Y 12 =-18.284X Asp -232.259X Cys +633.377X Met +15.533X Tyr +51.085X Pro +30.728X 白杨素 +0.145X 没食子酸 -0.466X 对香豆酸 +2.984X 芹菜素 +1.138X 迷迭香酸 +0.097X 香豆酸 +0.514X 丁香酸 -33.483X 柚皮素 -0.612X 原儿茶酸 +0.525X 阿魏酸 +0.222X 香草酸 -2.279X 反式肉桂酸 -0.006X 3,4-二甲基肉桂酸 -22.129X 咖啡酸苯乙基酯 +6.258X 犀草酸 -0.809X 异阿魏酸 -4.469。
[0068] Y 13 =9.697X Asp -314.135X Cys -327.65X Met +60.53X Tyr -10.139X Pro +3.076X 白杨素 -2.697X 没食子酸 +0.035X 对香豆酸 +3.185X 芹菜素 +0.97X 迷迭香酸 +0.31X 香豆酸 +0.044X丁香酸 -13.8X 柚皮素 +0.481X 原儿茶酸 +0.068X 阿魏酸 +0.063X 香草酸 +2.419X 反式肉桂酸 +0.201X 3,4-二甲基肉桂酸 +35.493X 咖啡酸苯乙基酯 +12.802X 犀草酸 -1.268X 异阿魏酸 -3.095.
[0069] Figure 6 This is a scatter plot of the classification of different types of honey provided in this embodiment.
[0070] Depend on Figure 6 The 14 honeys achieved good separation results. Only wolfberry honey, astragalus honey, chrysanthemum honey, and sophora flower honey showed some overlap. The remaining honeys were distinguishable and located in distinct spaces. To assess the reliability of the constructed model, this study employed a leave-one-out cross-validation strategy to systematically verify model performance.
[0071] Therefore, the linear discriminant analysis model of this embodiment can correctly classify 100% of the original grouped honey samples, with a cross-validation classification accuracy of 98.6%, of which 20% of chrysanthemum honey was misclassified as wolfberry honey. The discrimination and verification accuracy of the discriminant model are both high, indicating that the use of 21 differential components such as trans-cinnamic acid, protocatechuic acid, Pro, 3,4-dimethylcinnamic acid, rosmarinic acid, apigenin, Tyr, vanillic acid, coumaric acid, isoferulic acid, Cys, caffeic acid phenethyl ester, Met, Asp, gallic acid, naringenin, p-coumaric acid, ferulic acid, styrosinic acid, chrysin and syringic acid to discriminate different varieties of honey is relatively accurate, and can achieve honey variety discrimination to a certain extent. The honey variety classification function established based on the typical discriminant function is: Y 狼牙蜜 =1266.2X Asp +23329.8X Cys +26998.9X Met +9679.07X Tyr +8536.81X Pro +984.633X 白杨素 +73.037X 没食子酸 +4.72X 对香豆酸 -83.583X 芹菜素 +58.379X 迷迭香酸 +102.773X 香豆酸 -6.755X 丁香酸 +2465.22X 柚皮素 -32.185X<h2 style=";text-align:left;direction:ltr"> 原儿茶酸 <h2 style=";text-align:left;direction:ltr"> +16.12X<h2 style=";text-align:left;direction:ltr"> 阿魏酸 <h2 style=";text-align:left;direction:ltr"> +14.894X<h2 style=";text-align:left;direction:ltr"> 香草酸 <h2 style=";text-align:left;direction:ltr"> +23.68X<h2 style=";text-align:left;direction:ltr"> 反式肉桂酸 <h2 style=";text-align:left;direction:ltr"> -52.358X<h2 style=";text-align:left;direction:ltr"> 3,4-二甲基肉桂酸 <h2 style=";text-align:left;direction:ltr"> -1848.8X<h2 style=";text-align:left;direction:ltr"> 咖啡酸苯乙基酯 <h2 style=";text-align:left;direction:ltr"> +892.044X<h2 style=";text-align:left;direction:ltr"> 犀草酸 <h2 style=";text-align:left;direction:ltr"> +115.371X<h2 style=";text-align:left;direction:ltr"> 异阿魏酸 <h2 style=";text-align:left;direction:ltr"> -1625.6<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">
[0072] <h2 style=";text-align:left;direction:ltr"> Y<h2 style=";text-align:left;direction:ltr"> 槐花蜜 <h2 style=";text-align:left;direction:ltr"> =768.942X<h2 style=";text-align:left;direction:ltr"> Asp <h2 style=";text-align:left;direction:ltr"> +5488.662X<h2 style=";text-align:left;direction:ltr"> Cys <h2 style=";text-align:left;direction:ltr"> +2691.112X<h2 style=";text-align:left;direction:ltr"> Met <h2 style=";text-align:left;direction:ltr"> +6461.24X<h2 style=";text-align:left;direction:ltr"> Tyr <h2 style=";text-align:left;direction:ltr"> +2980.437X<h2 style=";text-align:left;direction:ltr"> Pro <h2 style=";text-align:left;direction:ltr"> -348.866X<h2 style=";text-align:left;direction:ltr"> 白杨素 <h2 style=";text-align:left;direction:ltr"> +47.618X<h2 style=";text-align:left;direction:ltr"> 没食子酸 <h2 style=";text-align:left;direction:ltr"> +21.036X<h2 style=";text-align:left;direction:ltr"> 对香豆酸 <h2 style=";text-align:left;direction:ltr"> -14.979X<h2 style=";text-align:left;direction:ltr"> 芹菜素 <h2 style=";text-align:left;direction:ltr"> +63.972X<h2 style=";text-align:left;direction:ltr"> 迷迭香酸 <h2 style=";text-align:left;direction:ltr"> +6.576X<h2 style=";text-align:left;direction:ltr"> 香豆酸 <h2 style=";text-align:left;direction:ltr"> +1.84X<h2 style=";text-align:left;direction:ltr"> 丁香酸 <h2 style=";text-align:left;direction:ltr"> +704.514X<h2 style=";text-align:left;direction:ltr"> 柚皮素 <h2 style=";text-align:left;direction:ltr"> -8.694X<h2 style=";text-align:left;direction:ltr"> 原儿茶酸 <h2 style=";text-align:left;direction:ltr"> +30.674X<h2 style=";text-align:left;direction:ltr"> 阿魏酸 <h2 style=";text-align:left;direction:ltr"> -7.067X<h2 style=";text-align:left;direction:ltr"> 香草酸 <h2 style=";text-align:left;direction:ltr"> +23.384X<h2 style=";text-align:left;direction:ltr"> 反式肉桂酸 <h2 style=";text-align:left;direction:ltr"> +59.906X<h2 style=";text-align:left;direction:ltr"> 3,4-二甲基肉桂酸 <h2 style=";text-align:left;direction:ltr"> -393.738X<h2 style=";text-align:left;direction:ltr"> 咖啡酸苯乙基酯 <h2 style=";text-align:left;direction:ltr"> +953.169X<h2 style=";text-align:left;direction:ltr"> 犀草酸 <h2 style=";text-align:left;direction:ltr"> +94.534X<h2 style=";text-align:left;direction:ltr"> 异阿魏酸 <h2 style=";text-align:left;direction:ltr"> -694,512.<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">
[0073] <h2 style=";text-align:left;direction:ltr"> Y<h2 style=";text-align:left;direction:ltr"> 枣花蜜 <h2 style=";text-align:left;direction:ltr"> =320.259X<h2 style=";text-align:left;direction:ltr"> Asp <h2 style=";text-align:left;direction:ltr"> +5671.554X<h2 style=";text-align:left;direction:ltr"> Cys <h2 style=";text-align:left;direction:ltr"> +19134.54X<h2 style=";text-align:left;direction:ltr"> Met <h2 style=";text-align:left;direction:ltr"> +7561.093X<h2 style=";text-align:left;direction:ltr"> Tyr <h2 style=";text-align:left;direction:ltr"> +4475.939X<h2 style=";text-align:left;direction:ltr"> Pro <h2 style=";text-align:left;direction:ltr"> -1404.86X<h2 style=";text-align:left;direction:ltr"> 白杨素 <h2 style=";text-align:left;direction:ltr"> +95.373X<h2 style=";text-align:left;direction:ltr"> 没食子酸 <h2 style=";text-align:left;direction:ltr"> +98.19X<h2 style=";text-align:left;direction:ltr"> 对香豆酸 <h2 style=";text-align:left;direction:ltr"> -78.82X<h2 style=";text-align:left;direction:ltr"> 芹菜素 <h2 style=";text-align:left;direction:ltr"> +65.441X<h2 style=";text-align:left;direction:ltr"> 迷迭香酸 <h2 style=";text-align:left;direction:ltr"> +11.557X<h2 style=";text-align:left;direction:ltr"> 香豆酸 <h2 style=";text-align:left;direction:ltr"> +39.107X<h2 style=";text-align:left;direction:ltr"> 丁香酸 <h2 style=";text-align:left;direction:ltr"> -1251.2X<h2 style=";text-align:left;direction:ltr"> 柚皮素 <h2 style=";text-align:left;direction:ltr"> -3.421X原儿茶酸 +66.943X 阿魏酸 -21.37X 香草酸 +4.848X 反式肉桂酸 +15.397X 3,4-二甲基肉桂酸 -1021.82X 咖啡酸苯乙基酯 +666.933X 犀草酸 +136.394X 异阿魏酸 -2496.28。
[0074] Y 油菜蜜 =157.093X Asp +840.842X Cys +2658.71X Met +2851.75X Tyr +479.635X Pro -1443X 白杨素 +42.411X 没食子酸 +4.788X 对香豆酸 +38.656X 芹菜素 +103.478X 迷迭香酸 -2.007X 香豆酸 +6.733X 丁香酸 -426.23X 柚皮素 +1.113X 原儿茶酸 +19.622X 阿魏酸 -3.72X 香草酸 -10.781X 反式肉桂酸 +24.621X 3,4-二甲基肉桂酸 -343.35X 咖啡酸苯乙基酯 +831.206X 犀草酸 +8.636X 异阿魏酸 -276.07。
[0075] Y 椴树蜜 =63.638X Asp +19731.72X Cys +24187.37X Met +4837.349X Tyr +1838.472X Pro -1195.53X 白杨素 +71.622X 没食子酸 +1.551X 对香豆酸 -49.946X 芹菜素 +69.514X 迷迭香酸 +23.559X 香豆酸 +1.946X 丁香酸 -201.934X 柚皮素 +5.245X 原儿茶酸+19.469X 阿魏酸 +1.371X 香草酸 -32.881X 反式肉桂酸 -7.247X 3,4-二甲基肉桂酸 -133.14X 咖啡酸苯乙基酯 +167.333X 犀草酸 +71.69X 异阿魏酸 -387.99。
[0076] Y 黄芪蜜 =701.824X Asp +31859.655X Cys +25027.24X Met +11653.202X Tyr +5442.434X Pro -839.17X 白杨素 +113.393X 没食子酸 -0.883X 对香豆酸 -8.111X 芹菜素 +54.274X 迷迭香酸 +105.289X 香豆酸 -2.702X 丁香酸 +896.13X 柚皮素 -15.738X 原儿茶酸 +28.731X 阿魏酸 +13.476X 香草酸 -56.339X 反式肉桂酸 -19.957X 3,4-二甲基肉桂酸 -2043.377X 咖啡酸苯乙基酯 +40.193X 犀草酸 +120.142X 异阿魏酸 -1135.493。
[0077] Y 枸杞蜜 =798.257X Asp +24536.03X Cys +14146.03X Met +6918.81X Tyr +7661.987X Pro -50.258X 白杨素 +64.798X 没食子酸 -12.777X 对香豆酸 -56.074X 芹菜素 +28.842X 迷迭香酸 +61.049X 香豆酸 +5.236X 丁香酸 +238.89X 柚皮素 -12.506X 原儿茶酸+38.402X 阿魏酸 +17.576X 香草酸 +110.885X 反式肉桂酸 -53.055X 3,4-二甲基肉桂酸 -1403.69X 咖啡酸苯乙基酯 +110.673X 犀草酸 +102.907X 异阿魏酸 -892.927。
[0078] Y 桂花蜜 =597.39X Asp +12992.19X Cys +20281.85X Met +4618.597X Tyr +4855.984X Pro +740.198X 白杨素 +67X 没食子酸 +1.531X 对香豆酸 -18.821X 芹菜素 +17.312X 迷迭香酸 +57.769X 香豆酸 +3.714X 丁香酸 -158.318X 柚皮素 -15.796X 原儿茶酸 +28.964X 阿魏酸 +0.832X 香草酸 -14.459X 反式肉桂酸 -33.669X 3,4-二甲基肉桂酸 -1206.98X 咖啡酸苯乙基酯 -93.412X 犀草酸 +90.529X 异阿魏酸 -413.46。
[0079] Y 茴香蜜 =1575.943X Asp +25537.33X Cys +6573.691X Met +9981.811X Tyr +12479.9X Pro -270.443X 白杨素 +48.144X 没食子酸 +27.36X 对香豆酸 -23.608X 芹菜素 +10.461X 迷迭香酸 +33.697X 香豆酸 +28.96X 丁香酸 -781.693X 柚皮素 -14.789X 原儿茶酸+28.318X 阿魏酸 -6.921X 香草酸 +61.039X 反式肉桂酸 -21.896X 3,4-二甲基肉桂酸 -326.014X 咖啡酸苯乙基酯 +382.268X 犀草酸 +166.932X 异阿魏酸 -1517.33。
[0080] Y 菊花蜜 =555.006X Asp +23250.447X Cys +9091.636X Met +4493.778X Tyr +7861.067X Pro -855.094X 白杨素 +43.549X 没食子酸 -18.237X 对香豆酸 -23.871X 芹菜素 +27.748X 迷迭香酸 +30.112X 香豆酸 +13.628X 丁香酸 -136.119X 柚皮素 +0.749X 原儿茶酸 +45.949X 阿魏酸 +23.693X 香草酸 +33.972X 反式肉桂酸 -40.284X 3,4-二甲基肉桂酸 -1079.592X 咖啡酸苯乙基酯 +93.912X 犀草酸 +80.063X 异阿魏酸 -780.464。
[0081] Y 荔枝蜜 =402.775X Asp +16652.12X Cys +15947.36X Met +10336.39X Tyr +4362.447X Pro +595.865X 白杨素 +0.351X 没食子酸 +7.134X 对香豆酸 +47.23X 芹菜素 +6.28X 迷迭香酸 +26.651X 香豆酸 +3.015X 丁香酸 +69.986X 柚皮素 -23.443X 原儿茶酸-17.27X 阿魏酸 -3.61X 香草酸 -5.91X 反式肉桂酸 +40.489X 3,4-二甲基肉桂酸 +459.097X 咖啡酸苯乙基酯 +137.573X 犀草酸 +85.037X 异阿魏酸 -575.268。
[0082] Y 荆条蜜 =672.746X Asp +15756.08X Cys +18901.09X Met +4857.928X Tyr +4442.072X Pro -152.494X 白杨素 +98.763X 没食子酸 -10.266X 对香豆酸 -69.637X 芹菜素 +27.215X 迷迭香酸 +24.85X 香豆酸 +31.253X 丁香酸 -1035.9X 柚皮素 +9.14X 原儿茶酸 +38.091X 阿魏酸 +5.319X 香草酸 -41.394X 反式肉桂酸 -26.98X 3,4-二甲基肉桂酸 -660.083X 咖啡酸苯乙基酯 +116.066X 犀草酸 +110.356X 异阿魏酸 -876.696。
[0083] Y 紫云英蜜 =274.629X Asp +9388.07X Cys +8440.59X Met +4136.71X Tyr +1313.95X Pro -1421.9X 白杨素 +138.338X 没食子酸 -9.809X 对香豆酸 -79.236X 芹菜素 +75.365X 迷迭香酸 +16.853X 香豆酸 +32.448X 丁香酸 +98.205X 柚皮素 -0.751X 原儿茶酸 +25.327X阿魏酸 +14.287X 香草酸 -67.503X 反式肉桂酸 -13.335X 3,4-二甲基肉桂酸 -558.1X 咖啡酸苯乙基酯 +808.246X 犀草酸 +42.132X 异阿魏酸 -720.05.
[0084] Y 枇杷蜜 =347.985X Asp +7945.84X Cys +1032.96X Met +2690.46X Tyr +2020.7X Pro -212.87X 白杨素 +10.995X 没食子酸 -3.817X 对香豆酸 -8.494X 芹菜素 +9.143X 迷迭香酸 +7.572X 香豆酸 +4.514X 丁香酸 -53.723X 柚皮素 -0.009X 原儿茶酸 +2.091X 阿魏酸 +3.455X 香草酸 +12.225X 反式肉桂酸 -3.763X 3,4-二甲基肉桂酸 +457.963X 咖啡酸苯乙基酯 +75.994X 犀草酸 +33.176X 异阿魏酸 -112.03.
[0085] Example 4 Random forests achieve prediction by combining multiple decision tree models. Each tree independently predicts the data, and the final result is determined through "voting" or averaging. Random forests are commonly used for classification and regression problems. By constructing random sample sets and feature sets, random forests reduce the possibility of overfitting. The parallel modeling of multiple decision trees achieves higher accuracy.
[0086] In this example, 21 characteristic difference components with VIP values greater than 1 in PLS-DA (trans-cinnamic acid, protocatechuic acid, Pro, 3,4-dimethylcinnamic acid, rosmarinic acid, apigenin, Tyr, vanillic acid, coumaric acid, isoferulic acid, Cys, caffeic acid phenethyl ester, Met, Asp, gallic acid, naringenin, p-coumaric acid, ferulic acid, styraxic acid, chrysin, and syringic acid) were selected as input, and honey varieties were selected as output. The sample data was randomly divided into a training set and a test set at a ratio of 8:2. Bootstrap sampling was used, the number of decision trees was set to 100, the minimum number of leaf node samples was 3, the training set data was used to complete the construction of the random forest model, and the test data was used for model prediction.
[0087] The model showed perfect prediction performance on both the training set and the test set, achieving 100% accuracy. Figures 7 to 9 .
[0088] Figure 7 This is the training set prediction accuracy graph of this embodiment; Figure 8 This is the prediction accuracy graph of the test set of this embodiment; Figure 9 This is the confusion matrix diagram of the training set of this embodiment; Figure 10 This is the confusion matrix diagram of the test set in this embodiment.
[0089] As can be seen from the above, the prediction accuracy of the random forest prediction model in this example was 100% for both the training set and the test set, indicating that it is feasible to use the amino acid and polyphenol compound content combined with the random forest algorithm to predict honey varieties.
[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for identifying honey varieties based on amino acids and polyphenolic compounds, characterized in that: The identification method comprises the following steps: (I) providing a honey sample to be tested, and determining the content of characteristic amino acids and characteristic polyphenolic compounds in the sample to be tested; The characteristic amino acids include Pro, Tyr, Cys, Met and Asp; the characteristic polyphenol compounds include trans-cinnamic acid, protocatechuic acid, 3,4-dimethylcinnamic acid, rosmarinic acid, apigenin, vanillic acid, coumaric acid, isoferulic acid, caffeic acid phenethyl ester, gallic acid, naringenin, p-coumaric acid, ferulic acid, styraxic acid, chrysin and syringic acid; (II) Constructing a honey variety identification model, introducing the contents of characteristic amino acids and characteristic polyphenolic compounds measured in the honey sample to be tested into the honey variety identification model, and identifying the honey variety indicating the honey sample to be tested.
2. The honey variety identification method according to claim 1, characterized in that: There are 14 honey varieties in step (II), including: wolfberry honey, locust flower honey, jujube flower honey, rapeseed honey, linden honey, astragalus honey, wolfberry honey, osmanthus honey, fennel honey, chrysanthemum honey, litchi honey, vitex honey, astragalus honey and loquat honey.
3. The method for identifying honey varieties according to claim 1 or 2, characterized in that: The method for constructing the honey variety identification model comprises: (A) Feature screening: S1: Collect and detect the data of amino acids and polyphenol compounds of 14 honey varieties as the first training dataset; The first training data set includes detection data of 17 amino acids and 24 polyphenolic compounds, and the 17 amino acids and 24 polyphenolic compounds are recorded as first quantitative species difference components; The 17 amino acids are Asp, Thr, Ser, Glu, Gly, Ala, Cys, Val, Met, Ile, Leu, Tyr, Phe, His, Lys, Arg, and Pro; The 24 polyphenolic compounds are chrysin, gallic acid, galangin, caffeic acid, p-coumaric acid, apigenin, rosmarinic acid, rutin, myricic acid, quercetin, coumaric acid, syringic acid, naringenin, morin, protocatechuic acid, squidrin, ferulic acid, vanillic acid, trans-cinnamic acid, 3,4-dimethylcinnamic acid, caffeic acid phenethyl ester, kaempferol, schizonepeta tenuifolia, and isoferulic acid; S2: Use partial least squares discriminant analysis to reduce the dimensionality of the first training data set to obtain the second training data set; The second training data set includes detection data of 5 amino acids and 16 polyphenolic compounds, and the 5 amino acids and 16 polyphenolic compounds are recorded as the second quantitative species difference components; The five amino acids are Pro, Tyr, Cys, Met and Asp; the 16 polyphenolic compounds are trans-cinnamic acid, protocatechuic acid, 3,4-dimethylcinnamic acid, rosmarinic acid, apigenin, vanillic acid, coumaric acid, isoferulic acid, caffeic acid phenethyl ester, gallic acid, naringenin, p-coumaric acid, ferulic acid, styrosinic acid, chrysin and syringic acid; (B) Model construction: Using the random forest algorithm, multiple decision trees are trained using the second training dataset to construct a random forest model as a honey variety identification model.
4. The method for identifying honey varieties according to claim 3, wherein: The dimensionality reduction processing method of step S2 includes: The first training data set was analyzed using the partial least squares discriminant analysis method, and the amino acid and polyphenol compound data with VIP values greater than 1 generated by the partial least squares discriminant analysis method were screened as characteristic difference component data to obtain the second training data set.
5. The honey variety identification method according to claim 3, characterized in that: In the model construction of step (B), each decision tree is trained by randomly selecting a sample data each time and selecting the characteristic difference component data of the second training data set with replacement, and the training is repeated multiple times by bootstrap resampling to obtain 100 decision trees, and the 100 decision trees are used to form a random forest model as a honey variety identification model; The honey variety identification model is used to evaluate the decision results of all decision trees, and the final identification results are output based on the majority principle.
6. The method for identifying honey varieties according to claim 3, wherein: In the model construction of step (B), for each decision tree, the second quantity-type difference component data is selected with replacement as a sample at the root node of the decision tree, all available second quantity-type difference components are used for evaluation when each node of the decision tree needs to be split, and the second quantity-type difference component that minimizes the node Gini coefficient is selected as the splitting condition for the node of the decision tree; According to the selected second number of type difference components, a splitting operation is performed on each node so that each child node contains a portion of the sample data; and this process is repeated until each leaf node contains at least 3 samples.
7. The method for identifying honey varieties according to claim 3, wherein: The sample data of multiple honey varieties obtained during data collection were divided into a certain proportion, of which 80% of the data was used as the training set and 20% of the data was used as the test set. Stratified sampling was used to ensure that each category had a reasonable proportion in the training and test sets. The synthetic minority oversampling technique was applied to the training set to address the class imbalance problem and maintain the original data distribution of the test set. The training set data is used for feature screening, model construction and optimization, and the constructed honey variety identification model is tested using the test set to evaluate the accuracy of the honey variety identification model on an independent data set.
8. The method for identifying honey varieties according to claim 3, wherein: The model construction in step (B) can be replaced by: The linear discriminant analysis method is used to establish a discriminant function using the second training data set as a training set, and a honey variety identification model is constructed.
9. Use of the honey variety identification method based on amino acids and polyphenolic compounds according to any one of claims 1 to 8 in honey variety identification.