Method and system for predicting flavone content of sunned green tea of ancient trees

Abstract

The invention provides a method and system for predicting the flavone content of sun-dried green tea of ancient trees in the technical field of tea tree planting management, and the method comprises the following steps: S1, recording the altitude of each soil sample and the age and height of a growing ancient tea tree, analyzing the soil parameters of each soil sample, and calculating the soil parameters of each soil sample; taking each soil parameter, altitude, tree age and tree height as influence factors; s2, labeling each influence factor to construct a data set; s3, performing regression analysis on the data set through an LASSO regression model to screen modeling factors; s4, performing multi-factor regression analysis on each modeling factor through a COX regression model; s5, creating a column graph model based on each modeling factor; s6, performing performance evaluation and stability evaluation on the column graph model; and S7, predicting the flavone content by using the column diagram model. The method has the advantage that the accuracy of predicting the flavone content of the sunned green tea of the ancient tree is greatly improved.

Description

Technical Field

[0001] The invention relates to the technical field of tea tree planting and management, and in particular to a method and system for predicting the flavonoid content of ancient tree sun-dried green tea. Background Art

[0002] Ancient tea trees refer to wild ancient tea trees and their communities distributed in natural forests, semi-domesticated artificially cultivated wild tea trees, and ancient tea gardens (forests) that have been artificially cultivated for more than 100 years. They have high economic and scientific research value. Among them, the tea made from the fresh leaves of ancient tea trees, following the procedures of withering, rolling, and sun drying, is called ancient tree sun-dried green tea.

[0003] Ancient tree sun-dried green tea is highly praised by the market for its harmonious taste, good resistance to brewing, and lasting sweetness. The content of flavonoids in ancient tree sun-dried green tea is generally 1.32% to 1.86%, and it has a soft astringent and bitter taste. It is an important component of green tea soup. Tea is the best source of dietary flavonoid compounds that cannot be synthesized by the human body. It has the effects of enzyme regulation, improving blood circulation, regulating cell apoptosis, preventing abnormal fat accumulation, anti-cancer, improving periodontal disease, anti-oxidation, weight loss, lowering blood sugar, inhibiting neurodegenerative diseases, and anti-Alzheimer's disease. With the in-depth research on flavonoids in recent years, it has been found that flavonoids are a promising scaffold for the development of new anti-dengue drugs. As a core structure, it has great potential for further development as a new anti-caking agent in the future. In addition, flavonoids can be used as a new multifunctional electrolyte additive to improve the cycle performance of high-voltage nickel manganese oxide lithium batteries. It can be seen that flavonoids can not only be widely used as natural antioxidants, natural pigments, natural food preservatives, etc. in the pharmaceutical, food and cosmetic industries, but also have great development potential in the industrial field. In order to increase the yield of flavonoids in ancient tree sun-dried green tea, there is a need to predict the flavonoid content of ancient tree sun-dried green tea, so as to optimize the planting environment and continuously increase the flavonoid yield.

[0004] However, since many factors affect the flavonoid content in ancient tree sun-dried green tea, the traditional method of predicting the flavonoid content of ancient tree sun-dried green tea based on planting experience often has unsatisfactory prediction accuracy. Therefore, how to provide a method and system for predicting the flavonoid content of ancient tree sun-dried green tea to improve the accuracy of predicting the flavonoid content of ancient tree sun-dried green tea has become a technical problem that needs to be solved urgently. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for predicting the flavonoid content of ancient tree sun-dried green tea, so as to improve the accuracy of predicting the flavonoid content of ancient tree sun-dried green tea.

[0006] In a first aspect, the present invention provides a method for predicting the flavonoid content of ancient tree sun-dried green tea, comprising the following steps:

[0007] Step S1, based on the five-point sampling method, soil sampling is performed on the planting site of ancient tree sun-drying green tea to obtain a plurality of soil samples, the altitude at which each soil sample is sampled and the age and height of the ancient tea trees growing thereon are recorded, the soil parameters of each soil sample are analyzed respectively, and the soil parameters, altitude, tree age and tree height are used as influencing factors;

[0008] Step S2: labeling each of the influencing factors to construct a data set, and performing a sample expansion operation on the data set;

[0009] Step S3, performing regression analysis on the data set by using a LASSO regression model, and screening out modeling factors from each of the soil components;

[0010] Step S4, performing multi-factor regression analysis on each of the modeling factors using a COX regression model;

[0011] Step S5, creating a nomogram model for predicting the flavonoid content of ancient tree sun-dried green tea based on each of the modeling factors;

[0012] Step S6, respectively performing performance evaluation and stability evaluation on the nomogram model through ROC curve and calibration curve, and deploying the nomogram model that passes the evaluation;

[0013] Step S7, obtaining a flavonoid content prediction request, verifying the flavonoid content prediction request and parsing it to obtain soil data, inputting the soil data into a nomogram model in real time to obtain a flavonoid content prediction result of ancient tree sun-dried green tea, and displaying the flavonoid content prediction result in the form of a chart through a visual interface.

[0014] Furthermore, the step S1 is specifically as follows:

[0015] Based on the five-point sampling method, the soil at the sunny and shady slopes of the ancient tree sun-dried green tea planting site was sampled at the top, middle and bottom of the slope after removing the surface 4-5 cm of soil, and the vertical depth of 20 cm and 40 cm of soil was sampled. Each sampling point was sampled 3 times to obtain several soil samples, and the altitude of each soil sample and the age and height of the ancient tea tree growing there were recorded;

[0016] The soil parameters of each soil sample, including at least ammonium nitrogen content, available phosphorus content, exchangeable potassium content, hydrolyzed nitrogen content, nitrate nitrogen content, organic carbon content, organic matter content, soil pH value, soil specific gravity, total nitrogen content, total phosphorus content, and total potassium content, are analyzed separately, and each soil parameter, altitude, tree age and tree height are taken as influencing factors.

[0017] Furthermore, the step S2 is specifically as follows:

[0018] The influencing factors were annotated to construct a data set, and the data set was expanded by using the Bootstrap method.

[0019] Furthermore, the step S3 is specifically as follows:

[0020] The data set was subjected to regression analysis by LASSO regression model, and several modeling factors significantly correlated with the flavonoid content of ancient tree sun-dried green tea were screened out from the influencing factors;

[0021] The loss function of the LASSO regression model is:

[0022]

[0023] Among them, J(θ) represents the loss value of the loss function; θ represents the sample number variable; m represents the number of samples; h θ represents the actual value of the i-th sample; x (i) represents the i-th sample; y (i) represents the predicted value of the i-th sample; λ represents the regularization parameter; n represents the number of features; θ j Represents the regression coefficient of the LASSO regression model.

[0024] Furthermore, the step S4 is specifically as follows:

[0025] A multi-factor regression analysis of the correlation between each modeling factor and the flavonoid content of ancient tree sun-dried green tea was performed using the COX regression model, and each modeling factor after the multi-factor regression analysis was further screened using the Akaike criterion;

[0026] The formula of the COX regression model is:

[0027] h(t)=h0(t)exp(β1x1+...+β j x j );

[0028] Among them, h(t) represents the risk function of the flavonoid content of ancient tree sun-dried green tea; h0(t) represents the intercept of the COX regression equation; x j represents the jth modeling factor; β j represents the jth regression coefficient;

[0029] The step S6 is specifically as follows:

[0030] The performance of the nomogram model is evaluated by the AUC value of the ROC curve, and the nomogram model is fitted by a calibration curve to perform stability evaluation. The nomogram model that passes the evaluation is then deployed, and an API interface called by the nomogram model is set.

[0031] In a second aspect, the present invention provides a system for predicting the flavonoid content of ancient tree sun-dried green tea, comprising the following modules:

[0032] The influencing factor collection module is used to obtain several soil samples from the planting area of ​​ancient tree sun-dried green tea based on the five-point sampling method, record the altitude of each soil sample and the age and height of the growing ancient tea tree, analyze the soil parameters of each soil sample respectively, and use the soil parameters, altitude, tree age and tree height as influencing factors;

[0033] A data set construction module, used to annotate each of the influencing factors to construct a data set, and perform a sample expansion operation on the data set;

[0034] A LASSO regression module, used to perform regression analysis on the data set through a LASSO regression model, and screen out modeling factors from each of the soil components;

[0035] A COX regression module, used for performing multi-factor regression analysis on each of the modeling factors through a COX regression model;

[0036] A nomogram model creation module, used to create a nomogram model for predicting the flavonoid content of ancient tree sun-dried green tea based on each of the modeling factors;

[0037] A nomogram model evaluation module, used to respectively evaluate the performance and stability of the nomogram model through ROC curve and calibration curve, and deploy the nomogram model that passes the evaluation;

[0038] The flavonoid content prediction module is used to obtain a flavonoid content prediction request, verify the flavonoid content prediction request and then analyze it to obtain soil data, input the soil data into the nomogram model in real time to obtain the flavonoid content prediction result of the ancient tree sun-dried green tea, and display the flavonoid content prediction result in the form of a chart through a visual interface.

[0039] Furthermore, the influencing factor collection module is specifically used for:

[0040] Based on the five-point sampling method, the soil at the sunny and shady slopes of the ancient tree sun-dried green tea planting site was sampled at the top, middle and bottom of the slope after removing the surface 4-5 cm of soil, and the vertical depth of 20 cm and 40 cm of soil was sampled. Each sampling point was sampled 3 times to obtain several soil samples, and the altitude of each soil sample and the age and height of the ancient tea tree growing there were recorded;

[0041] The soil parameters of each soil sample, including at least ammonium nitrogen content, available phosphorus content, exchangeable potassium content, hydrolyzed nitrogen content, nitrate nitrogen content, organic carbon content, organic matter content, soil pH value, soil specific gravity, total nitrogen content, total phosphorus content, and total potassium content, are analyzed separately, and each soil parameter, altitude, tree age and tree height are taken as influencing factors.

[0042] Furthermore, the data set construction module is specifically used for:

[0043] The influencing factors were annotated to construct a data set, and the data set was expanded by using the Bootstrap method.

[0044] Furthermore, the LASSO regression module is specifically used for:

[0045] The data set was subjected to regression analysis by LASSO regression model, and several modeling factors significantly correlated with the flavonoid content of ancient tree sun-dried green tea were screened out from the influencing factors;

[0046] The loss function of the LASSO regression model is:

[0047]

[0048] Among them, J(θ) represents the loss value of the loss function; θ represents the sample number variable; m represents the number of samples; h θ represents the actual value of the i-th sample; x (i) represents the i-th sample; y (i) represents the predicted value of the i-th sample; λ represents the regularization parameter; n represents the number of features; θ j Represents the regression coefficient of the LASSO regression model.

[0049] Furthermore, the COX regression module is specifically used for:

[0050] A multi-factor regression analysis of the correlation between each modeling factor and the flavonoid content of ancient tree sun-dried green tea was performed using the COX regression model, and each modeling factor after the multi-factor regression analysis was further screened using the Akaike criterion;

[0051] The formula of the COX regression model is:

[0052] h(t)=h0(t)exp(β1x1+...+β j x j );

[0053] Among them, h(t) represents the risk function of the flavonoid content of ancient tree sun-dried green tea; h0(t) represents the intercept of the COX regression equation; x j represents the jth modeling factor; βj represents the jth regression coefficient;

[0054] The nomogram model evaluation module is specifically used for:

[0055] The performance of the nomogram model is evaluated by the AUC value of the ROC curve, and the nomogram model is fitted by a calibration curve to perform stability evaluation. The nomogram model that passes the evaluation is then deployed, and an API interface called by the nomogram model is set.

[0056] The advantages of the present invention are:

[0057] 1. Several soil samples were obtained by sampling the soil at the planting site of ancient sun-dried green tea, and the altitude of each soil sample as well as the age and height of the ancient tea trees were recorded. The soil parameters of each soil sample were analyzed respectively, and each soil parameter, altitude, tree age and tree height were used as influencing factors. Then, each influencing factor was annotated to construct a data set and perform sample expansion operations. Then, the data set was regressed and analyzed using the LASSO regression model, and modeling factors were screened out from each soil component. Multifactor regression analysis was performed on each modeling factor using the COX regression model. Then, a nomogram model for predicting the flavonoid content of ancient sun-dried green tea was created based on each modeling factor. The performance and stability of the nomogram model were evaluated using the ROC curve and calibration curve, and the nomogram model that passed the evaluation was deployed. Finally, the nomogram model was obtained. Take a flavonoid content prediction request, verify the flavonoid content prediction request and then parse it to obtain soil data, input the soil data in real time into the nomogram model to obtain the flavonoid content prediction result of ancient tree sun-dried green tea, and display the flavonoid content prediction result in the form of a chart through a visual interface; that is, soil sampling and analysis are carried out on the planting area of ​​ancient tree sun-dried green tea to obtain the influencing factors of each soil sample, and then the LASSO regression model and COX regression model are used to screen out the influencing factors most relevant to the flavonoid content of ancient tree sun-dried green tea from various influencing factors as modeling factors, and a nomogram model is created based on each modeling factor. After performance and stability evaluation of the nomogram model, the nomogram model is used to predict the flavonoid content of ancient tree sun-dried green tea. This method combines machine learning and statistics, and ultimately greatly improves the accuracy of the prediction of flavonoid content in ancient tree sun-dried green tea.

[0058] 2. The nomogram model is used to predict the flavonoid content of ancient tree sun-dried green tea. Since the nomogram model is a very intuitive mathematical model, it can combine multiple factors to predict specific results. Compared with other prediction models, the nomogram model usually displays its results in the form of charts, which greatly improves the intuitiveness of the prediction results; and the nomogram model does not require a large amount of training data sets, which greatly improves the construction speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The present invention will be further described below in conjunction with embodiments with reference to the accompanying drawings.

[0060] Figure 1 It is a flow chart of a method for predicting the flavonoid content of ancient tree sun-dried green tea according to the present invention.

[0061] Figure 2 It is a structural schematic diagram of a system for predicting the flavonoid content of ancient tree sun-dried green tea according to the present invention. DETAILED DESCRIPTION

[0062] The technical solution in the embodiments of the present application has the following overall idea: soil sampling and analysis are carried out on the planting site of ancient tree sun-dried green tea to obtain the influencing factors of each soil sample, and then the influencing factors most relevant to the flavonoids content of ancient tree sun-dried green tea are screened out from various influencing factors through LASSO regression model and COX regression model as modeling factors, a nomogram model is created based on each modeling factor, and after the performance and stability evaluation of the nomogram model, the nomogram model is used to predict the flavonoids content of ancient tree sun-dried green tea, combining machine learning and statistics to improve the accuracy of the prediction of the flavonoids content of ancient tree sun-dried green tea.

[0063] Please refer to Figure 1 to Figure 2 As shown, a preferred embodiment of the method for predicting the flavonoid content of ancient tree sun-dried green tea of ​​the present invention comprises the following steps:

[0064] Step S1, based on the five-point sampling method, soil sampling is performed on the planting site of ancient tree sun-drying green tea to obtain a plurality of soil samples, the altitude at which each soil sample is sampled and the age and height of the ancient tea trees growing thereon are recorded, the soil parameters of each soil sample are analyzed respectively, and the soil parameters, altitude, tree age and tree height are used as influencing factors;

[0065] Step S2: labeling each of the influencing factors to construct a data set, and performing a sample expansion operation on the data set;

[0066] Step S3, performing regression analysis on the data set through the LASSO regression model, and screening out modeling factors from each of the soil components; by screening each of the influencing factors, the complexity of the nomogram model is effectively reduced, thereby obtaining better performance parameters and avoiding overfitting;

[0067] Step S4, performing multi-factor regression analysis on each of the modeling factors using a COX regression model;

[0068] Step S5, creating a nomogram model for predicting the flavonoid content of ancient tree sun-dried green tea based on each of the modeling factors;

[0069] The nomogram model is based on multi-factor regression analysis, integrating multiple modeling factors, and using line segments with scale lines to draw in proportion on the same plane to express the relationship between the modeling factors. By analyzing the size of the regression coefficient in the nomogram model, the influence of each modeling factor on the prediction of the flavonoid content of ancient tree sun-dried green tea can be obtained, and the corresponding score is assigned to each value of the dependent variable. The function conversion relationship between the total score and the flavonoid content of ancient tree sun-dried green tea is used to calculate the prediction probability of the flavonoid content of ancient tree sun-dried green tea.

[0070] Step S6, respectively performing performance evaluation and stability evaluation on the nomogram model through ROC curve and calibration curve, and deploying the nomogram model that passes the evaluation;

[0071] Step S7, obtaining a flavonoid content prediction request, verifying the flavonoid content prediction request and parsing it to obtain soil data, inputting the soil data into a nomogram model in real time to obtain a flavonoid content prediction result of ancient tree sun-dried green tea, and displaying the flavonoid content prediction result in the form of a chart through a visual interface.

[0072] The step S1 is specifically as follows:

[0073] Based on the five-point sampling method, the soil at the sunny and shady slopes of the ancient tree sun-dried green tea planting site is sampled at the top, middle and bottom sampling points of the slope after removing the surface 4-5 cm of soil, and the soil at a vertical depth of 20 cm and 40 cm is sampled. Each sampling point is sampled 3 times to obtain several soil samples, and the altitude of each soil sample and the age and height of the ancient tea tree growing thereon are recorded; in specific implementation, the sampling position, sampling depth and sampling weight must be consistent for each sampling, for example, 200g is sampled each time;

[0074] The soil parameters of each soil sample, including at least ammonium nitrogen content, available phosphorus content, exchangeable potassium content, hydrolyzed nitrogen content, nitrate nitrogen content, organic carbon content, organic matter content, soil pH value, soil specific gravity, total nitrogen content, total phosphorus content, and total potassium content, are analyzed separately, and each soil parameter, altitude, tree age and tree height are taken as influencing factors.

[0075] In specific implementation, the detection method of flavonoids in tea samples adopts aluminum chloride colorimetry; the determination of soil pH value adopts potentiometric method, the determination of organic matter content comes from the organic carbon content determined by potassium dichromate oxidation-spectrophotometry multiplied by a constant 1.724; the determination of soil specific gravity adopts density bottle method; the determination of total phosphorus content adopts alkali fusion-molybdenum antimony anti-spectrophotometry; the determination of effective phosphorus content adopts sodium bicarbonate extraction-molybdenum antimony anti-spectrophotometry; the determination of total potassium content adopts ICP-AES method, and the soil available potassium content is determined by extraction with neutral ferric acetate solution and flame photometry; the determination of total nitrogen content adopts Kjeldahl method; the determination of ammonium nitrogen content adopts combined extraction-colorimetric analysis method; the determination of nitrate nitrogen content adopts phenol disulfonic acid colorimetry; the determination of hydrolyzed nitrogen content is a determination method of hydrolyzing with sodium hydroxide solution, adding boric acid solution, and then titrating with standard acid.

[0076] The step S2 is specifically as follows:

[0077] The influencing factors were annotated to construct a data set, and the data set was expanded by using the Bootstrap method.

[0078] The basic idea of ​​the Bootstrap method is to extract several samples from the data set, conduct repeated experiments to construct multiple different data sets, and finally use the sample empirical distribution to replace the overall distribution; use the random replacement sampling method to perform corresponding operations, extract a certain number of samples from the data set, and then calculate the estimated statistics based on the extracted samples, and calculate the variance and distribution based on the calculation results.

[0079] The step S3 is specifically as follows:

[0080] The data set was subjected to regression analysis by LASSO (Least Absolute Shrinkage and Selection Operator) regression model, and several modeling factors significantly correlated with the flavonoid content of ancient tree sun-dried green tea were screened out from the influencing factors;

[0081] The loss function of the LASSO regression model is:

[0082]

[0083] Among them, J(θ) represents the loss value of the loss function; θ represents the sample number variable; m represents the number of samples; h θ represents the actual value of the i-th sample; x (i) represents the i-th sample; y (i) represents the predicted value of the i-th sample; λ represents the regularization parameter. The larger the λ value is, the fewer variables (modeling factors) are finally screened out by the LASSO regression model; n represents the number of features; θj Represents the regression coefficient of the LASSO regression model.

[0084] The LASSO regression model is a biased estimation method for processing data with complex collinearity by reducing the variable set and constructing a penalty function to compress the variable coefficients and make some regression coefficients zero, and retains the advantages of subset shrinkage.

[0085] In the coefficient distribution diagram of LASSO regression, as the horizontal axis (penalty coefficient) increases, the vertical axis (variable coefficient) will gradually be affected by the penalty term and compressed until it is finally compressed to 0 and stops changing. In order to reduce the risk of overfitting, the present invention selects the λ value at one standard error level for LASSO regression analysis.

[0086] The step S4 is specifically as follows:

[0087] A multi-factor regression analysis of the correlation between each modeling factor and the flavonoid content of ancient tree sun-dried green tea was performed using the COX regression model, and each modeling factor after the multi-factor regression analysis was further screened using the Akaike criterion;

[0088] The Akaike criterion is a comparative analysis through the AIC (Akaike Information Criterion) value. The AIC value is a standard for measuring the goodness of statistical model fitting. It is based on the concept of entropy and provides a standard for weighing the complexity of the estimated model and the goodness of fitting data. When selecting the best model from a set of available models, the model with the smallest AIC value is usually selected. The calculation formula of the AIC value is:

[0089] AIC = 2k-ln(L);

[0090] Among them, k represents the model parameter, that is, the modeling factor; L represents the likelihood function.

[0091] There are large differences between models, mainly reflected in the likelihood function term. When the likelihood function difference is not significant, the complexity of the model comes into play. When the model complexity increases (k increases), the likelihood function will also increase, and the AIC value will become smaller; when k is too large, the likelihood function growth rate slows down, resulting in an increase in the AIC value. Models that are too complex are prone to overfitting. The AIC value not only improves the model fit, but also introduces a penalty term to minimize the model parameters, which helps reduce the possibility of overfitting.

[0092] The formula of the COX regression model is:

[0093] h(t)=h0(t)exp(β1x1+...+β j x j );

[0094] Among them, h(t) represents the risk function of the flavonoid content of ancient tree sun-dried green tea; h0(t) represents the intercept of the COX regression equation; x j represents the jth modeling factor; β j represents the jth regression coefficient;

[0095] The step S6 is specifically as follows:

[0096] The performance of the nomogram model is evaluated by the AUC value of the ROC curve, and the nomogram model is fitted by a calibration curve to evaluate stability. The nomogram model that passes the evaluation is then deployed, and an API interface called by the nomogram model is set.

[0097] The ROC curve is used to evaluate the generalization performance of the model. It was originally used to detect radar signals in wars. It is currently mainly used to evaluate the performance of models generated by machine learning. The ROC curve can be used to comprehensively evaluate the advantages and disadvantages of different models, and the closer the curve is to the upper left, the better the prediction results. The ROC curve divides the final results into positive and negative. When the model accuracy is finally evaluated, the actual situation is divided into True Positive (TP), False Negative (FN), False Positive (FP), and True Negative (TN). When the predicted value is exactly the same as the true value, it is called True Positive or True Negative; when the predicted value is different from the true value, it is called False Negative or False Positive. The ordinate of the ROC curve is True Positive Chance (TPR). The larger the value, the more accurate the prediction. TRP = TP / (TP+FN). The abscissa of the ROC curve is False Positive Chance (FPR). The smaller the value, the more accurate the prediction. FRP = FP / (FP+TN).

[0098] Since the ROC curve cannot quantitatively evaluate the model performance, the model performance is evaluated by calculating the area under the curve (AUC). The AUC value represents the area under the ROC curve. The larger the AUC value, the stronger its classification ability.

[0099] The calibration curve is a tool to evaluate the degree of fit between the model prediction results and the actual situation. It can be used to evaluate the degree of fit between the nomogram and the changes in flavonoids content in ancient tree sun-dried green tea under abiotic stress.

[0100] The calibration curve is essentially a scatter plot that can be used to assess the agreement between the predicted and actual flavonoid contents. The calibration curve is a visualization of the Hosmer–Lemeshow goodness-of-fit test, comparing the predicted content with the actual content to determine if there is a significant difference. In the process of drawing the calibration curve, the predicted probabilities need to be binned under the guidance of uniformity and quantiles. Uniformity means that the width of each bin is the same, and quantile means that the number of data points in each bin is equal. Then the average of all sample predictions in each bin and the probability of a true positive in each bin are calculated respectively as the abscissa and ordinate of the scatter plot, and the calibration curve is created after connecting the points of the scatter plot. The long dashed line (ideal) of the calibration curve represents the ideal situation of the nomogram model, and the predicted values ​​are exactly the same as the actual values.

[0101] Logistic calibration represents the performance of the nomogram model corrected by the bootstrap method, while nonparametric represents the apparent accuracy of the nomogram model. If the predicted value is equal to the true outcome, the logistic calibration and nonparametric curves completely coincide; if the estimated value is higher than the true value, the nonparametric curve is below the logistic calibration; otherwise, the nonparametric curve is above the logistic calibration.

[0102] The step S7 is specifically as follows:

[0103] Obtain a flavonoid content prediction request carrying ciphertext data, a user ID, a timestamp, and a hash value, wherein the ciphertext data is obtained by encrypting the soil data using a 3DES algorithm, and the hash value is obtained by performing a hash calculation on the ciphertext data, the user ID, and the timestamp;

[0104] Parsing the received flavonoid content prediction request to obtain the ciphertext data, user ID, timestamp and hash value, performing integrity check on the ciphertext data, user ID and timestamp through the hash value, performing timeliness check through the timestamp, and then performing legitimacy check on the user ID through a preset user management table, and then decrypting the ciphertext data through the 3DES algorithm to obtain soil data;

[0105] The soil data is input into the nomogram model in real time to obtain the flavonoid content prediction result of the ancient tree sun-drying green tea, and the flavonoid content prediction result is displayed in the form of a chart through a visual interface.

[0106] By setting the flavonoid content prediction request to carry ciphertext data, user ID, timestamp and hash value, integrity verification can be performed based on the hash value, timeliness verification can be performed based on the timestamp, and legitimacy verification can be performed based on the user ID. In addition, the soil data can be prevented from being stolen in plain text during transmission. Four-fold security measures have been taken to greatly improve the security of calling the nomogram model.

[0107] A preferred embodiment of the system for predicting the flavonoid content of ancient tree sun-dried green tea of ​​the present invention comprises the following modules:

[0108] The influencing factor collection module is used to obtain several soil samples from the planting area of ​​ancient tree sun-dried green tea based on the five-point sampling method, record the altitude of each soil sample and the age and height of the growing ancient tea tree, analyze the soil parameters of each soil sample respectively, and use the soil parameters, altitude, tree age and tree height as influencing factors;

[0109] A data set construction module, used to annotate each of the influencing factors to construct a data set, and perform a sample expansion operation on the data set;

[0110] A LASSO regression module is used to perform regression analysis on the data set through a LASSO regression model, and screen out modeling factors from each of the soil components; by screening each of the influencing factors, the complexity of the nomogram model is effectively reduced, thereby obtaining better performance parameters and avoiding overfitting;

[0111] A COX regression module, used for performing multi-factor regression analysis on each of the modeling factors through a COX regression model;

[0112] A nomogram model creation module, used to create a nomogram model for predicting the flavonoid content of ancient tree sun-dried green tea based on each of the modeling factors;

[0113] The nomogram model is based on multi-factor regression analysis, integrating multiple modeling factors, and using line segments with scale lines to draw in proportion on the same plane to express the relationship between the modeling factors. By analyzing the size of the regression coefficient in the nomogram model, the influence of each modeling factor on the prediction of the flavonoid content of ancient tree sun-dried green tea can be obtained, and the corresponding score is assigned to each value of the dependent variable. The function conversion relationship between the total score and the flavonoid content of ancient tree sun-dried green tea is used to calculate the prediction probability of the flavonoid content of ancient tree sun-dried green tea.

[0114] A nomogram model evaluation module, used to respectively evaluate the performance and stability of the nomogram model through ROC curve and calibration curve, and deploy the nomogram model that passes the evaluation;

[0115] The flavonoid content prediction module is used to obtain a flavonoid content prediction request, verify the flavonoid content prediction request and then analyze it to obtain soil data, input the soil data into the nomogram model in real time to obtain the flavonoid content prediction result of the ancient tree sun-dried green tea, and display the flavonoid content prediction result in the form of a chart through a visual interface.

[0116] The influencing factor acquisition module is specifically used for:

[0117] Based on the five-point sampling method, the soil at the sunny and shady slopes of the ancient tree sun-dried green tea planting site is sampled at the top, middle and bottom sampling points of the slope after removing the surface 4-5 cm of soil, and the soil at a vertical depth of 20 cm and 40 cm is sampled. Each sampling point is sampled 3 times to obtain several soil samples, and the altitude of each soil sample and the age and height of the ancient tea tree growing thereon are recorded; in specific implementation, the sampling position, sampling depth and sampling weight must be consistent for each sampling, for example, 200g is sampled each time;

[0118] The soil parameters of each soil sample, including at least ammonium nitrogen content, available phosphorus content, exchangeable potassium content, hydrolyzed nitrogen content, nitrate nitrogen content, organic carbon content, organic matter content, soil pH value, soil specific gravity, total nitrogen content, total phosphorus content, and total potassium content, are analyzed separately, and each soil parameter, altitude, tree age and tree height are taken as influencing factors.

[0119] In specific implementation, the detection method of flavonoids in tea samples adopts aluminum chloride colorimetry; the determination of soil pH value adopts potentiometric method, the determination of organic matter content comes from the organic carbon content determined by potassium dichromate oxidation-spectrophotometry multiplied by a constant 1.724; the determination of soil specific gravity adopts density bottle method; the determination of total phosphorus content adopts alkali fusion-molybdenum antimony anti-spectrophotometry; the determination of effective phosphorus content adopts sodium bicarbonate extraction-molybdenum antimony anti-spectrophotometry; the determination of total potassium content adopts ICP-AES method, and the soil available potassium content is determined by extraction with neutral ferric acetate solution and flame photometry; the determination of total nitrogen content adopts Kjeldahl method; the determination of ammonium nitrogen content adopts combined extraction-colorimetric analysis method; the determination of nitrate nitrogen content adopts phenol disulfonic acid colorimetry; the determination of hydrolyzed nitrogen content is a determination method of hydrolyzing with sodium hydroxide solution, adding boric acid solution, and then titrating with standard acid.

[0120] The data set construction module is specifically used for:

[0121] The influencing factors were annotated to construct a data set, and the data set was expanded by using the Bootstrap method.

[0122] The basic idea of ​​the Bootstrap method is to extract several samples from the data set, conduct repeated experiments to construct multiple different data sets, and finally use the sample empirical distribution to replace the overall distribution; use the random replacement sampling method to perform corresponding operations, extract a certain number of samples from the data set, and then calculate the estimated statistics based on the extracted samples, and calculate the variance and distribution based on the calculation results.

[0123] The LASSO regression module is specifically used for:

[0124] The data set was subjected to regression analysis by LASSO (Least Absolute Shrinkage and Selection Operator) regression model, and several modeling factors significantly correlated with the flavonoid content of ancient tree sun-dried green tea were screened out from the influencing factors;

[0125] The loss function of the LASSO regression model is:

[0126]

[0127] Among them, J(θ) represents the loss value of the loss function; θ represents the sample number variable; m represents the number of samples; h θ represents the actual value of the i-th sample; x (i) represents the i-th sample; y (i) represents the predicted value of the i-th sample; λ represents the regularization parameter. The larger the λ value is, the fewer variables (modeling factors) are finally screened out by the LASSO regression model; n represents the number of features; θ j Represents the regression coefficient of the LASSO regression model.

[0128] The LASSO regression model is a biased estimation method for processing data with complex collinearity by reducing the variable set and constructing a penalty function to compress the variable coefficients and make some regression coefficients zero, and retains the advantages of subset shrinkage.

[0129] In the coefficient distribution diagram of LASSO regression, as the horizontal axis (penalty coefficient) increases, the vertical axis (variable coefficient) will gradually be affected by the penalty term and compressed until it is finally compressed to 0 and stops changing. In order to reduce the risk of overfitting, the present invention selects the λ value at one standard error level for LASSO regression analysis.

[0130] The COX regression module is specifically used for:

[0131] A multi-factor regression analysis of the correlation between each modeling factor and the flavonoid content of ancient tree sun-dried green tea was performed using the COX regression model, and each modeling factor after the multi-factor regression analysis was further screened using the Akaike criterion;

[0132] The Akaike criterion is to conduct comparative analysis through the AIC (Akaike Information Cr iter ion) value. The AIC value is a standard for measuring the goodness of statistical model fitting. It is based on the concept of entropy and provides a standard for weighing the complexity of the estimated model and the goodness of fitting data. When selecting the best model from a set of available models, the model with the smallest AIC value is usually selected. The calculation formula of the AIC value is:

[0133] AIC = 2k-ln(L);

[0134] Among them, k represents the model parameter, that is, the modeling factor; L represents the likelihood function.

[0135] There are large differences between models, mainly reflected in the likelihood function term. When the likelihood function difference is not significant, the complexity of the model comes into play. When the model complexity increases (k increases), the likelihood function will also increase, and the AIC value will become smaller; when k is too large, the likelihood function growth rate slows down, resulting in an increase in the AIC value. Models that are too complex are prone to overfitting. The AIC value not only improves the model fit, but also introduces a penalty term to minimize the model parameters, which helps reduce the possibility of overfitting.

[0136] The formula of the COX regression model is:

[0137] h(t)=h0(t)exp(β1x1+…+β j x j );

[0138] Among them, h(t) represents the risk function of the flavonoid content of ancient tree sun-dried green tea; h0(t) represents the intercept of the COX regression equation; x j represents the jth modeling factor; β j represents the jth regression coefficient;

[0139] The nomogram model evaluation module is specifically used for:

[0140] The performance of the nomogram model is evaluated by the AUC value of the ROC curve, and the nomogram model is fitted by a calibration curve to evaluate stability. The nomogram model that passes the evaluation is then deployed, and an API interface called by the nomogram model is set.

[0141] The ROC curve is used to evaluate the generalization performance of the model. It was originally used to detect radar signals in wars. It is currently mainly used to evaluate the performance of models generated by machine learning. The ROC curve can be used to comprehensively evaluate the advantages and disadvantages of different models, and the closer the curve is to the upper left, the better the prediction results. The ROC curve divides the final results into positive and negative. When the model accuracy is finally evaluated, the actual situation is divided into True Positive (TP), False Negative (FN), False Positive (FP), and True Negative (TN). When the predicted value is exactly the same as the true value, it is called True Positive or True Negative; when the predicted value is different from the true value, it is called False Negative or False Positive. The ordinate of the ROC curve is True Positive Chance (TPR). The larger the value, the more accurate the prediction. TRP = TP / (TP+FN). The abscissa of the ROC curve is False Positive Chance (FPR). The smaller the value, the more accurate the prediction. FRP = FP / (FP+TN).

[0142] Since the ROC curve cannot quantitatively evaluate the model performance, the model performance is evaluated by calculating the area under the curve (AUC). The AUC value represents the area under the ROC curve. The larger the AUC value, the stronger its classification ability.

[0143] The calibration curve is a tool to evaluate the degree of fit between the model prediction results and the actual situation. It can be used to evaluate the degree of fit between the nomogram and the changes in flavonoids content in ancient tree sun-dried green tea under abiotic stress.

[0144] The calibration curve is essentially a scatter plot that can be used to assess the agreement between the predicted and actual flavonoid contents. The calibration curve is a visualization of the Hosmer–Lemeshow goodness-of-fit test, comparing the predicted content with the actual content to determine if there is a significant difference. In the process of drawing the calibration curve, the predicted probabilities need to be binned under the guidance of uniformity and quantiles. Uniformity means that the width of each bin is the same, and quantile means that the number of data points in each bin is equal. Then the average of all sample predictions in each bin and the probability of a true positive in each bin are calculated respectively as the abscissa and ordinate of the scatter plot, and the calibration curve is created after connecting the points of the scatter plot. The long dashed line (ideal) of the calibration curve represents the ideal situation of the nomogram model, and the predicted values ​​are exactly the same as the actual values.

[0145] Logistic calibration represents the performance of the nomogram model corrected by the bootstrap method, while nonparametric represents the apparent accuracy of the nomogram model. If the predicted value is equal to the true outcome, the logistic calibration and nonparametric curves completely coincide; if the estimated value is higher than the true value, the nonparametric curve is below the logistic calibration; otherwise, the nonparametric curve is above the logistic calibration.

[0146] The flavonoid content prediction module is specifically used for:

[0147] Obtain a flavonoid content prediction request carrying ciphertext data, a user ID, a timestamp, and a hash value, wherein the ciphertext data is obtained by encrypting the soil data using a 3DES algorithm, and the hash value is obtained by performing a hash calculation on the ciphertext data, the user ID, and the timestamp;

[0148] Parsing the received flavonoid content prediction request to obtain the ciphertext data, user ID, timestamp and hash value, performing integrity check on the ciphertext data, user ID and timestamp through the hash value, performing timeliness check through the timestamp, and then performing legitimacy check on the user ID through a preset user management table, and then decrypting the ciphertext data through the 3DES algorithm to obtain soil data;

[0149] The soil data is input into the nomogram model in real time to obtain the flavonoid content prediction result of the ancient tree sun-drying green tea, and the flavonoid content prediction result is displayed in the form of a chart through a visual interface.

[0150] By setting the flavonoid content prediction request to carry ciphertext data, user ID, timestamp and hash value, integrity verification can be performed based on the hash value, timeliness verification can be performed based on the timestamp, and legitimacy verification can be performed based on the user ID. In addition, the soil data can be prevented from being stolen in plain text during transmission. Four-fold security measures have been taken to greatly improve the security of calling the nomogram model.

[0151] In summary, the advantages of the present invention are:

[0152] 1. Several soil samples were obtained by sampling the soil at the planting site of ancient sun-dried green tea, and the altitude of each soil sample as well as the age and height of the ancient tea trees were recorded. The soil parameters of each soil sample were analyzed respectively, and each soil parameter, altitude, tree age and tree height were used as influencing factors. Then, each influencing factor was annotated to construct a data set and perform sample expansion operations. Then, the data set was regressed and analyzed using the LASSO regression model, and modeling factors were screened out from each soil component. Multifactor regression analysis was performed on each modeling factor using the COX regression model. Then, a nomogram model for predicting the flavonoid content of ancient sun-dried green tea was created based on each modeling factor. The performance and stability of the nomogram model were evaluated using the ROC curve and calibration curve, and the nomogram model that passed the evaluation was deployed. Finally, the nomogram model was obtained. Take a flavonoid content prediction request, verify the flavonoid content prediction request and then parse it to obtain soil data, input the soil data in real time into the nomogram model to obtain the flavonoid content prediction result of ancient tree sun-dried green tea, and display the flavonoid content prediction result in the form of a chart through a visual interface; that is, soil sampling and analysis are carried out on the planting area of ​​ancient tree sun-dried green tea to obtain the influencing factors of each soil sample, and then the LASSO regression model and COX regression model are used to screen out the influencing factors most relevant to the flavonoid content of ancient tree sun-dried green tea from various influencing factors as modeling factors, and a nomogram model is created based on each modeling factor. After performance and stability evaluation of the nomogram model, the nomogram model is used to predict the flavonoid content of ancient tree sun-dried green tea. This method combines machine learning and statistics, and ultimately greatly improves the accuracy of the prediction of flavonoid content in ancient tree sun-dried green tea.

[0153] 2. The nomogram model is used to predict the flavonoid content of ancient tree sun-dried green tea. Since the nomogram model is a very intuitive mathematical model, it can combine multiple factors to predict specific results. Compared with other prediction models, the nomogram model usually displays its results in the form of charts, which greatly improves the intuitiveness of the prediction results; and the nomogram model does not require a large amount of training data sets, which greatly improves the construction speed.

[0154] Although the specific implementation modes of the present invention are described above, those skilled in the art should understand that the specific implementation modes described are only illustrative and are not intended to limit the scope of the present invention. Equivalent modifications and changes made by those skilled in the art in accordance with the spirit of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A method for predicting the flavonoid content of ancient tree sun-dried green tea, characterized in that: The steps include: Step S1, based on the five-point sampling method, soil sampling is performed on the planting site of ancient tree sun-drying green tea to obtain a plurality of soil samples, the altitude at which each soil sample is sampled and the age and height of the ancient tea trees growing thereon are recorded, the soil parameters of each soil sample are analyzed respectively, and the soil parameters, altitude, tree age and tree height are used as influencing factors; Step S2: labeling each of the influencing factors to construct a data set, and performing a sample expansion operation on the data set; Step S3, performing regression analysis on the data set by using a LASSO regression model, and screening out modeling factors from each of the soil components; Step S4, performing multi-factor regression analysis on each of the modeling factors using a COX regression model; Step S5, creating a nomogram model for predicting the flavonoid content of ancient tree sun-dried green tea based on each of the modeling factors; Step S6, respectively performing performance evaluation and stability evaluation on the nomogram model through ROC curve and calibration curve, and deploying the nomogram model that passes the evaluation; Step S7, obtaining a flavonoid content prediction request, verifying the flavonoid content prediction request and parsing it to obtain soil data, inputting the soil data into a nomogram model in real time to obtain a flavonoid content prediction result of ancient tree sun-dried green tea, and displaying the flavonoid content prediction result in the form of a chart through a visual interface.

2. The method for predicting the flavonoid content of ancient tree sun-dried green tea according to claim 1, characterized in that: The step S1 is specifically as follows: Based on the five-point sampling method, the soil at the sunny and shady slopes of the ancient tree sun-dried green tea planting site was sampled at the top, middle and bottom of the slope after removing the surface 4-5 cm of soil, and the vertical depth of 20 cm and 40 cm of soil was sampled. Each sampling point was sampled 3 times to obtain several soil samples, and the altitude of each soil sample and the age and height of the ancient tea tree growing there were recorded; The soil parameters of each soil sample, including at least ammonium nitrogen content, available phosphorus content, exchangeable potassium content, hydrolyzed nitrogen content, nitrate nitrogen content, organic carbon content, organic matter content, soil pH value, soil specific gravity, total nitrogen content, total phosphorus content, and total potassium content, are analyzed separately, and each soil parameter, altitude, tree age and tree height are taken as influencing factors.

3. The method for predicting the flavonoid content of ancient tree sun-dried green tea according to claim 1, characterized in that: The step S2 is specifically as follows: The influencing factors were annotated to construct a data set, and the data set was expanded by using the Bootstrap method.

4. The method for predicting the flavonoid content of ancient tree sun-dried green tea according to claim 1, characterized in that: The step S3 is specifically as follows: The data set was subjected to regression analysis by LASSO regression model, and several modeling factors significantly correlated with the flavonoid content of ancient tree sun-dried green tea were screened out from the influencing factors; The loss function of the LASSO regression model is: Among them, J(θ) represents the loss value of the loss function; θ represents the sample number variable; m represents the number of samples; h θ represents the actual value of the i-th sample; x (i) represents the i-th sample; y (i) represents the predicted value of the i-th sample; λ represents the regularization parameter; n represents the number of features; θ j Represents the regression coefficient of the LASSO regression model.

5. The method for predicting the flavonoid content of ancient tree sun-dried green tea according to claim 1, characterized in that: The step S4 is specifically as follows: A multi-factor regression analysis of the correlation between each modeling factor and the flavonoid content of ancient tree sun-dried green tea was performed using the COX regression model, and each modeling factor after the multi-factor regression analysis was further screened using the Akaike criterion; The formula of the COX regression model is: h(t)=h0(t)exp(β1x1+...+β j x j ); Among them, h(t) represents the risk function of the flavonoid content of ancient tree sun-dried green tea; h0(t) represents the intercept of the COX regression equation; x j represents the jth modeling factor; β j represents the jth regression coefficient; The step S6 is specifically as follows: The performance of the nomogram model is evaluated by the AUC value of the ROC curve, and the nomogram model is fitted by a calibration curve to perform stability evaluation. The nomogram model that passes the evaluation is then deployed, and an API interface called by the nomogram model is set.

6. A system for predicting the flavonoid content of ancient tree sun-dried green tea, characterized in that: Includes the following modules: The influencing factor collection module is used to obtain several soil samples from the planting area of ​​ancient tree sun-dried green tea based on the five-point sampling method, record the altitude of each soil sample and the age and height of the growing ancient tea tree, analyze the soil parameters of each soil sample respectively, and use the soil parameters, altitude, tree age and tree height as influencing factors; A data set construction module, used to annotate each of the influencing factors to construct a data set, and perform a sample expansion operation on the data set; A LASSO regression module, used to perform regression analysis on the data set through a LASSO regression model, and screen out modeling factors from each of the soil components; A COX regression module, used for performing multi-factor regression analysis on each of the modeling factors through a COX regression model; A nomogram model creation module, used to create a nomogram model for predicting the flavonoid content of ancient tree sun-dried green tea based on each of the modeling factors; A nomogram model evaluation module, used to respectively evaluate the performance and stability of the nomogram model through ROC curve and calibration curve, and deploy the nomogram model that passes the evaluation; The flavonoid content prediction module is used to obtain a flavonoid content prediction request, verify the flavonoid content prediction request and then analyze it to obtain soil data, input the soil data into the nomogram model in real time to obtain the flavonoid content prediction result of the ancient tree sun-dried green tea, and display the flavonoid content prediction result in the form of a chart through a visual interface.

7. The system for predicting flavonoid content of ancient tree sun-dried green tea according to claim 6, characterized in that: The influencing factor acquisition module is specifically used for: Based on the five-point sampling method, the soil at the sunny and shady slopes of the ancient tree sun-dried green tea planting site was sampled at the top, middle and bottom of the slope after removing the surface 4-5 cm of soil, and the vertical depth of 20 cm and 40 cm of soil was sampled. Each sampling point was sampled 3 times to obtain several soil samples, and the altitude of each soil sample and the age and height of the ancient tea tree growing there were recorded; The soil parameters of each soil sample, including at least ammonium nitrogen content, available phosphorus content, exchangeable potassium content, hydrolyzed nitrogen content, nitrate nitrogen content, organic carbon content, organic matter content, soil pH value, soil specific gravity, total nitrogen content, total phosphorus content, and total potassium content, are analyzed separately, and each soil parameter, altitude, tree age and tree height are taken as influencing factors.

8. The system for predicting flavonoid content of ancient tree sun-dried green tea according to claim 6, characterized in that: The data set construction module is specifically used for: The influencing factors were annotated to construct a data set, and the data set was expanded by using the Bootstrap method.

9. The system for predicting flavonoid content of ancient tree sun-dried green tea according to claim 6, characterized in that: The LASSO regression module is specifically used for: The data set was subjected to regression analysis by LASSO regression model, and several modeling factors significantly correlated with the flavonoid content of ancient tree sun-dried green tea were screened out from the influencing factors; The loss function of the LASSO regression model is: Among them, J(θ) represents the loss value of the loss function; θ represents the sample number variable; m represents the number of samples; h θ represents the actual value of the i-th sample; x (i) represents the i-th sample; y (i) represents the predicted value of the i-th sample; λ represents the regularization parameter; n represents the number of features; θ j Represents the regression coefficient of the LASSO regression model.

10. The system for predicting flavonoid content of ancient tree sun-dried green tea according to claim 6, characterized in that: The COX regression module is specifically used for: A multi-factor regression analysis of the correlation between each modeling factor and the flavonoid content of ancient tree sun-dried green tea was performed using the COX regression model, and each modeling factor after the multi-factor regression analysis was further screened using the Akaike criterion; The formula of the COX regression model is: h(t)=h0(t)exp(β1x1+...+β j x j ); Among them, h(t) represents the risk function of the flavonoid content of ancient tree sun-dried green tea; h0(t) represents the intercept of the COX regression equation; x j represents the jth modeling factor; β j represents the jth regression coefficient; The nomogram model evaluation module is specifically used for: The performance of the nomogram model is evaluated by the AUC value of the ROC curve, and the nomogram model is fitted by a calibration curve to perform stability evaluation. The nomogram model that passes the evaluation is then deployed, and an API interface called by the nomogram model is set.

Images