Sea cucumber soaking method and soaking state prediction method

Abstract

The invention relates to the technical field of marine product processing and application, and discloses a sea cucumber soaking method and a soaking state prediction method. Spreading the selected dried sea cucumbers and steaming for 30 minutes; soaking the steamed dried sea cucumbers in an ice-water mixture for 20 hours; cleaning the sea cucumbers soaked for 20 hours, removing internal organs, and cooking for 3 hours; and putting the sea cucumbers boiled for 3 hours into the ice-water mixture, and continuously soaking for 48 hours. The soaking state prediction method comprises the following steps: acquiring sea cucumber samples in different soaking states, measuring state indexes, and inputting the state indexes into a pre-constructed XgBoost model for prediction to obtain a sea cucumber soaking state prediction result. According to the method, each soaking link can be better controlled, the problem of unstable product quality is effectively solved, the soaking state of the sea cucumber is more objectively and accurately predicted, and the prediction accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of seafood processing and application, and more specifically, to a sea cucumber soaking method and a soaking state prediction method. Background Art

[0002] As a marine food with high nutritional value, sea cucumbers are usually processed into dried products for easy storage and transportation. Before consumption, dried sea cucumbers need to go through a soaking process to restore their original shape and taste. The soaking quality of sea cucumbers directly affects their edible quality and economic value. Therefore, developing a scientific and controllable sea cucumber soaking method and a technology to accurately predict the soaking state are of great significance to improving the quality of sea cucumber products.

[0003] The traditional method of soaking sea cucumber mainly depends on the experience of production staff, and the soaking degree is judged by sensory evaluation. Specifically, the production staff judges whether the sea cucumber has reached the soaking standard of each stage by touch, observation and other methods based on experience. However, this method has the problem that it is highly subjective and the soaking degree of different batches of sea cucumbers is difficult to control to be consistent. At the same time, there is a lack of systematic research on the changes of various parameters in the sea cucumber soaking process in the prior art, and the soaking state cannot be accurately predicted. Therefore, the existing sea cucumber soaking technology has the defects of unstable soaking quality and the inability to accurately predict the soaking state. Summary of the invention

[0004] In order to overcome the defects of the sea cucumber soaking technology, such as unstable soaking quality and inability to accurately predict the soaking state, the present invention proposes the following technical solutions:

[0005] In a first aspect, the present invention provides a method for soaking sea cucumbers, comprising:

[0006] Select dried sea cucumber;

[0007] Lay out the selected dried sea cucumber and steam for 30 minutes;

[0008] After steaming for 30 minutes, soak the dried sea cucumber in an ice-water mixture for 20 hours, and then wash and remove the internal organs;

[0009] Boil the cleaned and gutted sea cucumber for 3 hours;

[0010] After boiling for 3 hours, place the sea cucumber into the ice water mixture and continue soaking for 48 hours.

[0011] In a second aspect, the present invention provides a method for predicting the foaming state of sea cucumbers, comprising:

[0012] The sea cucumber soaking method as described in the first aspect is used to obtain sea cucumber samples in different soaking states.

[0013] Determine the status index of the sea cucumber sample.

[0014] The state index is input into the pre-built XgBoost model for prediction to obtain the prediction result of the foaming state of the sea cucumber.

[0015] As a preferred technical solution, the state indicators of the sea cucumber are measured after being soaked for 20 hours and cleaned to remove the internal organs, after being boiled for 3 hours, and after being soaked for 48 hours.

[0016] As a preferred technical solution, the state indicators include the weight, length, texture characteristics and moisture state of the sea cucumber.

[0017] As a preferred technical solution, the moisture state includes the moisture content of the sea cucumber and the law of moisture change.

[0018] As a preferred technical solution, the water content variation law includes the variation law of the bound water ratio, semi-bound water ratio and free-flowing water ratio of the sea cucumber.

[0019] As a preferred technical solution, before inputting the state indicator into the pre-built XgBoost model for prediction, the method further includes:

[0020] Perform at least one of the following processing on the status indicator data:

[0021] Delete duplicate data.

[0022] Delete missing values ​​or fill in the mean.

[0023] Delete or replace outliers.

[0024] Normalize or standardize the status indicator data.

[0025] As a preferred technical solution, when constructing the XgBoost model, the method further includes:

[0026] The Pearson correlation coefficient matrix and Spearman correlation coefficient matrix of each status indicator were calculated respectively.

[0027] The state indicators in which the correlation coefficients of the Pearson correlation coefficient matrix and the Spearman correlation coefficient matrix are both greater than m or less than -m are used as the prediction result parameters of the XgBoost model.

[0028] As a preferred technical solution, after obtaining the prediction result parameters, the method further includes:

[0029] The XgBoost model is simulated by a single missing experiment using the prediction parameters. When the fit of the XgBoost model remains unchanged or increases, the current prediction parameters are deleted. When the fit of the XgBoost model decreases, the current prediction parameters are retained as the prediction result parameters of the XgBoost model.

[0030] As a preferred technical solution, before inputting the state indicator into the pre-built XgBoost model for prediction, the method also includes using a grid search method to find and optimize the best parameters of the XgBoost model.

[0031] The beneficial effects of the present invention include at least:

[0032] (1) The present invention first steams the dried sea cucumber for 30 minutes to soften the tissue structure of the sea cucumber and prepare for subsequent soaking. Then, the sea cucumber steamed for 30 minutes is placed in an ice-water mixture for soaking for 20 hours, which can promote the sea cucumber to quickly absorb water and swell, and restore the shape and texture similar to that of fresh sea cucumber. The soaked sea cucumber is then boiled for 3 hours to further soften the sea cucumber tissue, making it more tender and having a better taste. Finally, the sea cucumber boiled for 3 hours is continued to be soaked for 48 hours to fully absorb water and swell, and finally achieve an ideal soaking effect. Compared with the traditional soaking method relying on manual experience, this scientific and systematic method can better control each soaking link, so that the shape and taste of the sea cucumber are fully restored, thereby effectively solving the problem of unstable product quality.

[0033] (2) The present invention obtains sea cucumber samples in different swelling states, then measures a series of state indicators of these sea cucumber samples and inputs them into a pre-built XgBoost machine learning model for training and prediction. This prediction method based on machine learning can predict the swelling state of sea cucumbers more objectively and accurately than manual subjective judgment through data-driven automated prediction, thereby improving the accuracy of prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a schematic diagram of the process of the sea cucumber soaking method provided in Example 1.

[0035] Figure 2 The present invention is a schematic flow chart of the method for soaking sea cucumbers provided in Example 2.

[0036] Figure 3 (a) is a moisture analysis diagram of the first soaking of the sea cucumber provided in Example 2, Figure 3 (b) is the final soaking moisture analysis diagram of the sea cucumber provided in Example 2.

[0037] Figure 4(a) is a schematic diagram of the Spearman correlation matrix provided in Example 2, Figure 4 (b) is a schematic diagram of the Person correlation matrix provided in Example 2. DETAILED DESCRIPTION

[0038] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred technical solutions. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred technical solutions are only for illustrating the present invention, not for limiting the scope of protection of the present invention.

[0039] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0040] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0041] Example 1

[0042] This embodiment provides a method for soaking sea cucumbers. Figure 1 As shown, Figure 1 A schematic flow chart of a method for soaking sea cucumbers provided in this embodiment, the method comprising the following steps:

[0043] Select dried sea cucumbers.

[0044] In this embodiment, the selected sea cucumber weighs 5 to 6 g and is produced in Dalian, Liaoning Province. It has no damage on the appearance and is free of diseases and insect pests.

[0045] Lay the selected dried sea cucumbers flat and steam them for 30 minutes;

[0046] After steaming for 30 minutes, soak the dried sea cucumber in an ice-water mixture for 20 hours, and then wash and remove the internal organs;

[0047] Put the cleaned and gutted sea cucumber into boiling water and cook for 3 hours;

[0048] After boiling for 3 hours, place the sea cucumber into the ice water mixture and continue soaking for 48 hours.

[0049] It is understandable that the present embodiment firstly softens the tissue structure of the sea cucumber by steaming the dry sea cucumber for 30 minutes, so as to prepare for subsequent soaking. Then, the sea cucumber steamed for 30 minutes is placed in an ice-water mixture for soaking for 20 hours, which can promote the rapid water absorption and expansion of the sea cucumber, and restore the shape and texture of the fresh sea cucumber. Then, the sea cucumber after soaking is boiled for 3 hours, which can further soften the sea cucumber tissue, make it more tender, and have a better taste. Finally, the sea cucumber boiled for 3 hours is continued to be soaked for 48 hours, so that it fully absorbs water and expands, and finally reaches the desired soaking effect. Compared with the traditional soaking method relying on artificial experience, this scientific and systematic method can better control each soaking link, so that the shape and taste of the sea cucumber are fully restored, thereby effectively solving the problem of unstable product quality.

[0050] Example 2

[0051] like Figure 2 As shown, this embodiment proposes a method for predicting the foaming state of sea cucumbers, such as Figure 2 As shown, Figure 2 The present invention is a schematic flow chart of a method for predicting the foaming state of sea cucumbers provided in this embodiment, wherein the method comprises the following steps:

[0052] Using the sea cucumber soaking method described in Example 1, obtaining sea cucumber samples in different soaking states;

[0053] Determine the state index of the sea cucumber sample;

[0054] The state index is input into the pre-built XgBoost model for prediction to obtain the prediction result of the foaming state of the sea cucumber.

[0055] In this embodiment, the state indicators of the sea cucumber were measured after being soaked for 20 hours and cleaned to remove the internal organs, after being boiled for 3 hours, and after being soaked for 48 hours.

[0056] In this embodiment, the state indicators include the weight, length, texture characteristics and moisture state of the sea cucumber.

[0057] In this embodiment, the moisture state includes the moisture content of the sea cucumber and the law of moisture change.

[0058] In this embodiment, the water content variation pattern includes the variation pattern of bound water ratio, semi-bound water ratio and free-flowing water ratio of sea cucumber.

[0059] As an exemplary description, this embodiment collects data on the weight, length, texture characteristics and moisture status of 839 sea cucumbers during different soaking processes, constructs a production database of different sea cucumber soaking processes, and uses the measured indicators to predict the sensory state of the soaked sea cucumbers.

[0060] Table 1 Sensory evaluation table of sea cucumber soaking

[0061]

[0062]

[0063] Table 2 Data statistics during the sea cucumber soaking process

[0064]

[0065] As shown in Tables 1 and 2, the key parameters and processing control points in the foaming process were determined by combining the experimental results with the sensory results:

[0066] The key points of the first foaming:

[0067] -Weight distribution is about 2 times that of the original sea cucumber.

[0068] -The moisture content is controlled between 55-60%.

[0069] -The foaming ratio is 1:2.

[0070] - Critical free-flowing water content is 61.99 ± 9.38%.

[0071] The key points of the second foaming:

[0072] -The weight distribution is about 4 times that of the original sea cucumber.

[0073] -The moisture content is controlled between 80-85%.

[0074] -The foaming ratio is 1:4.8-5.0.

[0075] - Critical free-flowing water content is 95.39 ± 0.57%.

[0076] By precisely controlling these key parameters, such as weight distribution, moisture content, swelling ratio, and free-flowing water content, the swelling process of sea cucumbers can be effectively guided to ensure the quality stability and sensory quality of the final product. Controlling these key process control points is crucial to improving the quality and shelf life of sea cucumber products.

[0077] In this embodiment, before inputting the state indicator into the pre-built XgBoost model for prediction, the method further includes:

[0078] Perform at least one of the following processing on the status indicator data:

[0079] Delete duplicate data;

[0080] Delete missing values ​​or fill in the mean;

[0081] Delete or replace outliers;

[0082] Normalize or standardize the status indicator data.

[0083] In this embodiment, when constructing the XgBoost model, the method further includes:

[0084] Calculate the Pearson correlation coefficient matrix and Spearman correlation coefficient matrix for each status indicator respectively;

[0085] The state indicators in which the correlation coefficients of the Pearson correlation coefficient matrix and the Spearman correlation coefficient matrix are both greater than m or less than -m are used as the prediction result parameters of the XgBoost model.

[0086] In this embodiment, after obtaining the prediction result parameters, the method further includes:

[0087] The XgBoost model is simulated by a single missing experiment using the prediction parameters. When the fit of the XgBoost model remains unchanged or increases, the current prediction parameters are deleted. When the fit of the XgBoost model decreases, the current prediction parameters are retained as the prediction result parameters of the XgBoost model.

[0088] As an exemplary illustration, Figure 3 and Figure 4 As shown in the figure, through correlation heat map analysis, appropriate factors can be selected as prediction parameters. Figure 3 (a) is the moisture analysis diagram of sea cucumber during the first soaking. Figure 3 (b) is the final soaking moisture analysis diagram of sea cucumber, Figure 3 (a) and Figure 3Each curve in (b) represents the change in signal intensity of the same batch of sea cucumber samples under different conditions. As the soaking process proceeds, the bound water in the sea cucumber gradually transforms into semi-bound water and free-flowing water, resulting in a gradual increase in signal intensity. In the initial state, the signal intensity of the sea cucumber is low, reflecting its water state mainly in the form of bound water. As the soaking time increases, the increase in signal intensity indicates the change in water state, that is, the water is gradually released from the bound form inside the tissue to a more free-flowing form. Finally, after sufficient soaking, the signal intensity reaches the highest, indicating that most of the water in the sea cucumber is already free-flowing water and the sea cucumber has reached a good soaking state.

[0089] Figure 4 (a) is the schematic diagram of the Spearman correlation matrix. Figure 4 (b) is a schematic diagram of the Person correlation matrix, which shows the correlation analysis between multiple parameters in the sea cucumber soaking process. The statistical relationship between these parameters is revealed by the Spearman and Pearson correlation matrices. The Spearman correlation matrix evaluates the monotonic relationship between parameters and is applicable to non-normally distributed data, while the Pearson correlation matrix measures the linear relationship between parameters and is applicable to normally distributed data. Through these two analyses, we can identify parameters that are critical to the prediction of the sea cucumber soaking state, such as weight, length, moisture content, and hardness. The correlation analysis results of these parameters are crucial for building the XgBoost model because they help us optimize the model input and improve the prediction accuracy. Through the Pearson and Spearman correlation matrices, it can be found that the moisture change index A4 has the highest correlation during the sea cucumber soaking process. At the same time, factors such as weight, length, soaking ratio, and moisture content are relatively consistent with the sensory perception, while hardness shows a significant negative correlation. The above factors with correlations greater than 0.8 or less than -0.8 are used as candidate key parameters for prediction. It should be noted that there is also a strong correlation between these candidate parameters, and there may be a certain possibility of weight increase.

[0090] Next, the actual weight, length, swelling ratio, moisture content, moisture change A4 and hardness, which are parameters with strong correlation, are used as the initial prediction parameters of the model to calculate the degree of fit and overfitting of the model. On this basis, the full parameters are first used for simulation, and then a single missing experiment is conducted based on the strength of the correlation between the factors and the increase in the actual weight. If the degree of fit of the model remains unchanged or increases, the current prediction parameter is deleted; if the degree of fit of the model decreases, the current prediction parameter is retained. Finally, the key prediction parameters used in model construction are determined to be: weight, moisture content, free flowing water A4 and hardness of sea cucumbers. The selection of these parameters helps to improve the prediction performance of the model and provide a valuable reference for the optimization of the sea cucumber swelling process.

[0091] During the prediction process, the XgBoost model can predict the quality indicators of sea cucumbers, such as the degree of swelling, taste and texture, based on these prediction parameters. Specifically, the weight of the sea cucumber reflects its degree of swelling, the water content determines the taste and texture, the free-flowing water A4 ratio affects the texture and taste, and hardness is an important indicator to measure the swelling state. The XgBoost model will first input the state indicators of the sea cucumber, such as weight, water content, free-flowing water A4 and hardness. After data preprocessing, the model prediction is performed based on the learned rules, and finally the prediction results of the sea cucumber's degree of swelling, taste and texture are output.

[0092] In this embodiment, before inputting the state indicator into the pre-built XgBoost model for prediction, the method also includes using a grid search method to find and optimize the optimal parameters of the XgBoost model.

[0093] Understandably, XgBoost is an optimized distributed gradient boosting library that is particularly suitable for processing tabular data. GridSearchCV is a function in the Scikit-learn library that is used to exhaustively search a given parameter grid to find the best parameter combination. The main optimized hyperparameters include: max_depth (maximum depth of the tree), min_child_weight (minimum sum of weights required in a node), gamma (minimum loss reduction for controlling whether to further split a node), subsample (the proportion of subsamples used to train each tree to the total training set), colsample_bytree (the proportion of features used to train each tree to the total features), eta (learning rate), lambda (coefficient of the L2 regularization term), and alpha (coefficient of the L1 regularization term).

[0094] As an exemplary explanation, the machine learning model used in this embodiment is the XgBoost model, and Python 3.10 is used for analysis and calculations are performed on a Windows 10 64-bit system. The data package uses tools such as Pandas, Sklearn, and XgBoost.

[0095] In the modeling process, weight change, length change, moisture content, swelling ratio, free water ratio and hardness were first used as predictors, and the sensory evaluation of sea cucumbers was used as the output result. The original data set was then divided into training and test subsets of randomly selected records (80% and 20%, respectively), and the training hyperparameters were optimized and tested. Finally, 10-fold cross-difference validation was used to prevent overfitting and selection bias problems. This modeling method helps to fully explore the potential relationship between multiple key indicators and the sensory quality of sea cucumbers, and improves the reliability of predictions through model optimization and cross-validation, providing a more accurate basis for the optimization of sea cucumber soaking processing technology.

[0096] The relevant code description of XgBoost model hyperparameters is as follows:

[0097] import XgBoost as xgb

[0098] from sklearn.model_selection import GridSearchCV

[0099] param_grid = {

[0100] 'max_depth': [1, 3, 5, 7, 10],

[0101] 'min_child_weight': [1, 2, 3, 4, 5],

[0102] 'gamma': [0, 0.01, 0.1, 0.2],

[0103] 'subsample': [0.5, 0.6, 0.7, 0.8, 0.9, 1.0],

[0104] 'colsample_bytree': [0.5, 0.6, 0.7, 0.8, 0.9, 1.0],

[0105] 'eta': [0.01, 0.1, 0.2],

[0106] 'lambda': [1, 1.25, 1.5, 1.75, 2],

[0107] 'alpha': [0, 0.1, 0.5]

[0108] }

[0109] xgb_model=xgb.XGBRegressor()

[0110] grid_search=GridSearchCV(estimator=xgb_model, param_grid=param_grid, cv=10, scoring='neg_mean_squared_error') grid_search.fit(X_train, y_train)

[0111] print("Best parameters found:",grid_search.best_params_)

[0112] Through the above settings, the best optimization parameters are determined as:

[0113] 'max_depth' is 7; 'min_child_weight': is 2, 'gamma': is 0.2, 'subsample': is 1, 'colsample_bytree': is 0.8, 'eta': is 0.01, 'lambda': is 1, 'alpha': is 0.5.

[0114] Based on the optimization results of the above parameters, the test set was further used for testing. The accuracy of the prediction set using the XgBoost model was 88%, indicating that 88% of all predicted samples were correctly classified. In terms of precision and recall, the model has a precision of 90% and a recall of 85% on the positive class, which shows that the model has high accuracy in identifying positive samples and can also capture most of the positive samples. The F1 score is 0.87, which is the harmonic mean of precision and recall, and comprehensively reflects the accuracy and robustness of the model. The logarithmic loss value is 0.35, a low value that indicates that the model has high accuracy in output probability estimation. The R^2 (coefficient of determination) value is 0.81, indicating that the model can explain 81% of the variability of the dependent variable and the model fits well.

[0115] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0116] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "N" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0117] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0118] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array, a field programmable gate array, etc.

[0119] A person skilled in the art may understand that all or part of the steps in the above-mentioned embodiment method may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0120] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for soaking sea cucumbers, characterized in that: The following steps are involved: Select dried sea cucumber; Lay out the selected dried sea cucumber and steam for 30 minutes; After steaming for 30 minutes, soak the dried sea cucumber in an ice-water mixture for 20 hours, and then wash and remove the internal organs; Boil the cleaned and gutted sea cucumber for 3 hours; After boiling for 3 hours, place the sea cucumber into the ice water mixture and continue soaking for 48 hours.

2. A method for predicting the foaming state of sea cucumbers, characterized in that: include: Using the sea cucumber soaking method as claimed in claim 1, obtaining sea cucumber samples in different soaking states; Determine the state index of the sea cucumber sample; The state index is input into the pre-built XgBoost model for prediction to obtain the prediction result of the foaming state of the sea cucumber.

3. The method for predicting the foaming state of sea cucumber according to claim 2, wherein: The status indicators of sea cucumbers were measured after they were soaked for 20 hours and cleaned to remove the internal organs, boiled for 3 hours, and soaked for 48 hours.

4. The method for predicting the foaming state of sea cucumber according to claim 2, characterized in that: The state indicators include the weight, length, texture characteristics and moisture state of the sea cucumber.

5. The method for predicting the foaming state of sea cucumber according to claim 4, characterized in that: The moisture state includes the moisture content of the sea cucumber and the law of moisture change.

6. The method for predicting the foaming state of sea cucumber according to claim 5, characterized in that: The water content variation rules include the variation rules of bound water ratio, semi-bound water ratio and free-flowing water ratio of sea cucumbers.

7. The method for predicting the foaming state of sea cucumber according to claim 2, characterized in that: Before inputting the state indicator into the pre-built XgBoost model for prediction, the method further includes: Perform at least one of the following processing on the status indicator data: Delete duplicate data; Delete missing values ​​or fill in the mean; Delete or replace outliers; Normalize or standardize the status indicator data.

8. The method for predicting the foaming state of sea cucumber according to claim 2, characterized in that: When constructing the XgBoost model, the method further includes: Calculate the Pearson correlation coefficient matrix and Spearman correlation coefficient matrix for each status indicator respectively; The state indicators in which the correlation coefficients of the Pearson correlation coefficient matrix and the Spearman correlation coefficient matrix are both greater than m or less than -m are used as the prediction result parameters of the XgBoost model.

9. The method for predicting the foaming state of sea cucumber according to claim 8, characterized in that: After obtaining the prediction result parameters, the method further includes: The XgBoost model is simulated by a single missing experiment using the prediction parameters. When the fit of the XgBoost model remains unchanged or increases, the current prediction parameters are deleted. When the fit of the XgBoost model decreases, the current prediction parameters are retained as the prediction result parameters of the XgBoost model.

10. The method for predicting the foaming state of sea cucumber according to claim 2, characterized in that: Before inputting the state indicator into the pre-built XgBoost model for prediction, the method also includes using a grid search method to find and optimize the best parameters of the XgBoost model.

Images